# Closefinch: full site text > Closefinch is a hosted CRM and agent platform. It captures a team's email, calendar, meetings and Slack into accounts, contacts and deals, and runs AI agents that prepare, follow up and forecast. Outgoing actions wait for a person's approval by default. Generated from the pages listed in https://www.closefinch.com/llms.txt. --- Source: https://www.closefinch.com/ # Closefinch > Closefinch is a hosted CRM and agent platform. It captures a team's email, calendar, meetings and Slack into accounts, contacts and deals, and runs AI agents that prepare, follow up and forecast. Outgoing actions wait for a person's approval by default. ## A CRM that fills itself in, and agents that ask before they act Closefinch captures your team's email, calendar, meetings and Slack into accounts, contacts and deals. Then it runs agents that prepare, follow up and forecast. Anything that leaves the building waits for a person's approval, unless you say otherwise. - Free for up to 3 seats, no card - Team $35 and Pro $79 per seat a month, billed annually - Prices public in HTML, Markdown and JSON Start free: https://app.closefinch.com/sign-up · Sign in: https://app.closefinch.com/sign-in · Pricing: https://www.closefinch.com/pricing ## Records that fill themselves in Each teammate connects their mailbox and calendar once. Business conversations are filed to the right account and deal, and personal and marketing mail is left out. - Gmail, Microsoft 365, Slack and calls from your phone system - Fields suggested from what buyers wrote, each with the sentence it came from - Keep your current CRM and sync both ways, or move over ![A demo account's page: its details on the left and, on the right, a timeline of the emails exchanged with its contacts](https://www.closefinch.com/shots/account-light.png) [More about capture](https://www.closefinch.com/product/crm) ## Meetings that turn into follow-ups Transcripts come from Zoom, Google Meet and Microsoft Teams, or from a recorder bot that joins from the calendar. - A live transcript that turns promises into follow-ups - A summary, tasks and a drafted follow-up email after each call - Recording rules you set: external meetings only, or only when invited ![A demo team's meetings: each call's name, start and end, video link and owner](https://www.closefinch.com/shots/meetings-light.png) [More about meetings](https://www.closefinch.com/product/meetings) ## Agents that ask before they act Start from a template or describe the job. Agents run on a schedule or when something happens, and by default anything that leaves the building waits in an approval inbox. - Autonomy levels you choose, per agent - Every step in a trace marked read, reason or write - Spending budgets that pause runs when they're reached ![The agent template gallery: built-in agents such as a stalled deal nudge and a meeting prep brief, each with what it does and a button to use it](https://www.closefinch.com/shots/templates-light.png) [More about agents](https://www.closefinch.com/product/agents) ## Outreach written for each person Define who to contact and who never to contact, then run sequences that draft each message from research. Do-not-contact rules and sending limits apply to every send. - Multi-step sequences with reply detection - Replies sorted by intent, with drafts a person approves - Your own Gmail or Outlook mailboxes as senders ![The prospecting page, where a team defines its ideal customer profile: who it may contact and who it must never contact](https://www.closefinch.com/shots/prospecting-light.png) [More about outbound](https://www.closefinch.com/product/outbound) ## Forecasts built from what buyers said Risk flags, next steps and a forecast where each commit deal has to meet six conditions in the buyer's own words. - Deal risk flags that clear when the buyer replies - Roll-ups with manager overrides and submitted calls - Coverage, slipped deals, stage velocity and win rates ![Part of a demo team's forecast: the roll-up by person, and the deals behind it with their amount, stage and forecast category](https://www.closefinch.com/shots/forecast-light.png) [More about deals and forecasts](https://www.closefinch.com/product/deals) ## Works with the tools your team already uses Google, Microsoft 365, Slack, Zoom, Google Meet, Microsoft Teams, Salesforce, Aircall, Dialpad, Zendesk, Intercom, Stripe, Snowflake, Notion, and more: [all integrations](https://www.closefinch.com/integrations). ## Built for agents as well as people Everything in the app is in the API. Assistants connect through the hosted MCP server, and scripts use the CLI. - REST API with OpenAPI 3.1 and scoped keys - MCP server for Claude, ChatGPT, Cursor and other clients - Webhooks in and out; outbound deliveries are signed - llms.txt and a Markdown version of every page ```bash # Connect Claude Code to your CRM claude mcp add --transport http revenueagent \ https://api.closefinch.com/mcp \ --header "Authorization: Bearer $RA_API_KEY" # Or call the REST API curl -s "https://api.closefinch.com/v1/deals?limit=5" \ -H "Authorization: Bearer $RA_API_KEY" ``` [Read the docs](https://www.closefinch.com/docs) ## Security you can check Each organization's data is kept apart, owners use two-step sign-in, and every change by a person or an agent is in the audit log. - Connected-account credentials stored encrypted - Hosted in Germany - No selling of data and no model training on it - A published subprocessor list and DPA. No SOC 2 report yet; we say so [How we protect your data](https://www.closefinch.com/security) Closefinch is a service of Ask Lou LLC, a Delaware limited liability company. --- Source: https://www.closefinch.com/product # A tour of the product The main jobs it does for a revenue team, each shown in the app with fictional demo data. Every area after the tour has its own page listing what's built. ## Records that fill themselves in Each teammate connects their mailbox and calendar once. Business conversations are filed to the right account and deal, and personal and marketing mail is left out. - Gmail, Microsoft 365, Slack and calls from your phone system - Fields suggested from what buyers wrote, each with the sentence it came from - Keep your current CRM and sync both ways, or move over ![A demo account's page: its details on the left and, on the right, a timeline of the emails exchanged with its contacts](https://www.closefinch.com/shots/account-light.png) Everything in this area: [CRM and data capture](https://www.closefinch.com/product/crm) ## Meetings that turn into follow-ups Transcripts come from Zoom, Google Meet and Microsoft Teams, or from a recorder bot that joins from the calendar. - A live transcript that turns promises into follow-ups - A summary, tasks and a drafted follow-up email after each call - Recording rules you set: external meetings only, or only when invited ![A demo team's meetings: each call's name, start and end, video link and owner](https://www.closefinch.com/shots/meetings-light.png) Everything in this area: [Meetings](https://www.closefinch.com/product/meetings) ## Agents that ask before they act Start from a template or describe the job. Agents run on a schedule or when something happens, and by default anything that leaves the building waits in an approval inbox. - Autonomy levels you choose, per agent - Every step in a trace marked read, reason or write - Spending budgets that pause runs when they're reached ![The agent template gallery: built-in agents such as a stalled deal nudge and a meeting prep brief, each with what it does and a button to use it](https://www.closefinch.com/shots/templates-light.png) Everything in this area: [Agents](https://www.closefinch.com/product/agents) ## Outreach written for each person Define who to contact and who never to contact, then run sequences that draft each message from research. Do-not-contact rules and sending limits apply to every send. - Multi-step sequences with reply detection - Replies sorted by intent, with drafts a person approves - Your own Gmail or Outlook mailboxes as senders ![The prospecting page, where a team defines its ideal customer profile: who it may contact and who it must never contact](https://www.closefinch.com/shots/prospecting-light.png) Everything in this area: [Prospecting and outbound](https://www.closefinch.com/product/outbound) ## Forecasts built from what buyers said Risk flags, next steps and a forecast where each commit deal has to meet six conditions in the buyer's own words. - Deal risk flags that clear when the buyer replies - Roll-ups with manager overrides and submitted calls - Coverage, slipped deals, stage velocity and win rates ![Part of a demo team's forecast: the roll-up by person, and the deals behind it with their amount, stage and forecast category](https://www.closefinch.com/shots/forecast-light.png) Everything in this area: [Deals and forecasting](https://www.closefinch.com/product/deals) ## Every area Each area's page lists what's built there, and notes anything that's off by default or needs your own account with another service. | Area | What it covers | |---|---| | [CRM and data capture](https://www.closefinch.com/product/crm) | Accounts, contacts, deals, leads, tasks and notes, with custom fields and objects. Records fill in from connected email, calendars, meetings, Slack and other tools, and filled-in values link back to where they came from. Keep your current CRM and sync with it, or move over. | | [Research and signals](https://www.closefinch.com/product/research) | Account and person research with cited sources, enrichment that fills only empty fields, and signals that say why an account is worth a look now. Facts quote their source; what the agent infers is labelled as inference. | | [Prospecting and outbound](https://www.closefinch.com/product/outbound) | ICP rules, lists and scoring, then sequences that draft each message from research, and an inbox for the replies. Do-not-contact rules, sending limits and deliverability checks apply to every send. | | [Inbound and scheduling](https://www.closefinch.com/product/inbound) | A chat agent for your website, forms, lead qualification and routing, and a scheduler that books meetings on your team's calendars. | | [Meetings](https://www.closefinch.com/product/meetings) | Transcripts from Zoom, Google Meet and Microsoft Teams or from the recorder bot, notes and follow-ups after each call, and prompts during calls for people who opt in. | | [Deals and forecasting](https://www.closefinch.com/product/deals) | Deal risk, next steps, coaching scorecards and forecasts, built from what buyers said on calls and in email. | | [Voice, SMS and messaging](https://www.closefinch.com/product/voice) | AI phone agents for inbound and outbound calls, two-way SMS, and WhatsApp, Telegram, Instagram and iMessage channels. Calls and texts run on your own Telnyx account, and outbound AI calling is off by default. | | [Agents](https://www.closefinch.com/product/agents) | Agents built from templates or from a goal you describe. They run on a schedule, on an event or on demand, and show every step they took. Their writes go through the policy engine before they happen. | | [Quotes and RFPs](https://www.closefinch.com/product/quotes) | Quotes drafted from a deal description with your products, prices and approval rules, and questionnaire answers from a reviewed answer library. | | [Customer success](https://www.closefinch.com/product/customers) | Renewals, churn signals, onboarding handoffs, surveys and collections. Outreach is drafted for a person to approve. | | [Analytics](https://www.closefinch.com/product/analytics) | Questions answered in plain English with citations, dashboards, campaign and team views, attribution and exports. | | [Security, admin and compliance](https://www.closefinch.com/product/security) | Roles and fine-grained access, an audit log of every action, two-step sign-in and passkeys, scoped API keys, and consent and opt-out records across email, SMS and calls. | | [Apps, API and MCP](https://www.closefinch.com/product/platform) | The web app, the Slack app and a mobile Today view, plus a REST API, a hosted MCP server, a CLI and webhooks for building on top. | | [Industry packs](https://www.closefinch.com/product/industries) | Packs for industrial distribution, equipment dealers and rental, consumer goods, home services and fundraising. Each adds its own data, triggers and agents. | What's new: the [changelog](https://www.closefinch.com/changelog). What it costs: [pricing](https://www.closefinch.com/pricing). The tools it connects to: [integrations](https://www.closefinch.com/integrations). --- Source: https://www.closefinch.com/product/crm # CRM and data capture Accounts, contacts, deals, leads, tasks and notes, with custom fields and objects. Records fill in from connected email, calendars, meetings, Slack and other tools, and filled-in values link back to where they came from. Keep your current CRM and sync with it, or move over. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Records, fields and CRM sync - Native CRM objects (Accounts, Contacts, Deals/Opportunities, Leads, Tasks, Notes, Activities, Meetings, Lists) with a REST schema endpoint per resource. - Typed custom fields (short/long text, number, boolean, date, URL, single/multi-select, single/multi-user, single/multi-record, currency, percentage), archivable, org-level creation toggle. - Custom objects and arbitrary relationships between objects. - Multiple pipelines (sales, partnerships, renewals); colored stages with an immutable stage type (start, in-progress, won, lost) that drives reporting and signals; Kanban board. - Lead object with lifecycle (new → contacted → qualified/unqualified → converted), source metadata and estimated value; conversion creates Contact, Account and optional Deal. - Saved and dynamic lists with multi-membership, archive and restore, and auto-refresh as ICP data changes. - Custom task statuses with category (To Do / In Progress / Done) and color. - Append-only, source-backed activity timeline: synced items update or disappear when the source changes; custom events posted with `external_id` for idempotency. - Business or knowledge graph linking accounts, people, conversations, signals and their relationships, used as shared agent memory. - Object history showing how each value changed over time, with point-in-time reads through the API. - Account memory: facts that people and agents save about an account, each with its evidence and revision, readable through the API and MCP. - Auto-create Accounts by email-domain matching and Contacts from email, calendar attendees, Slack Connect members and notetaker attendees. - Duplicate detection and merge into a survivor record, with redirect semantics for API clients (301 on read, 410 on write). - A dedup check before an agent creates or updates a record, as an agent tool. - Dismiss or ignore noise objects so they stop resurfacing. - AI-defined field: user names the field, the LLM picks the type, writes the extraction instructions and autofills it. - Backfill a new or changed field across all historical records. - Computed, calculated and conditional fields. - Schema-less, conversation-first capture where structured fields are derived from unstructured conversation data. - Signals as account fields: score, status, signals, and the latest signal with its date and source. - AI-proposed account matching with source and confidence: a person approves a match once and later emails and imports follow it. - Canonical master data plus a business glossary/ontology (terms, metric definitions, calendar rules, lineage) read by every connected AI tool. - Email aliases: users verify alternate addresses (up to 10) so activity attributes correctly. - Ownership model: deals inherit the account owner, `owner:none` filtering, plain-English bulk reassignment. - Agent-driven migration from Salesforce, HubSpot, Attio, Pipedrive, Close, Zoho, Affinity, DealCloud or spreadsheets: CSV upload, field-map confirmation, dedup and pipeline import. - Bulk-create API/CLI with idempotent row keys and bulk operations. - Run beside your current CRM with two-way Salesforce, HubSpot, Attio or Pipedrive sync, and migrate in phases by object, pipeline and team. - Overlay mode: agents write through to your existing CRM instead of replacing it, and agent changes wait for a person's approval before they reach it. - Write-through to Close, Zoho CRM, Affinity and DealCloud (Pipedrive also syncs both ways). - CRM hygiene agent: checks custom fields, finds duplicates, missing fields, broken associations, stale owners and zombie deals, ranks them by impact, and applies only the fixes a person approves, with a recovery trail. - Sandbox environments for testing schema and agent changes. - Two-way sync conflict resolution: field-level precedence rules, a conflict review UI and published sync latency. - Per-field value provenance: every auto-filled CRM value links to the exact email, transcript span or event that produced it. - Relationship-strength field per contact (who on the team knows them, how, last two-way touch) usable by agents and gates. ## Capture from email, calendar and other tools - Gmail sync every minute with a qualification filter: business mail is kept, and personal, marketing and automated mail is left out unless it matches the CRM. - Outlook / Microsoft 365 mail sync. - Google Calendar sync from the calendars you pick, by live event-watch and a one-minute sweep; private and declined events are left out. - Outlook Calendar sync. - Slack channel and thread capture linked to the right account or deal. - Slack Connect: each shared channel maps to one account; create channels with a prefix or link existing ones with history backfill; external members become contacts; channel health shown. - WhatsApp, Telegram and Instagram message capture into the graph. *(Needs your own WhatsApp Business, Telegram or Instagram account.)* - Notetaker imports from Granola, Notion meeting notes, Circleback, Fireflies, Fathom, Grain, Gong and Chorus, and a signed webhook for any other notetaker. - Dialer and phone-call capture (Aircall, Dialpad, RingCentral, Zoom Phone, Orum, Outreach). - In-person capture via a mobile recorder or hardware (Plaud), with a first-use consent screen. - Billing events as account activity (Stripe webhook matched by customer email; read-only Stripe key for former-customer detection). - Product-analytics event ingestion (PostHog, Segment, Amplitude, Mixpanel, or a first-party JS/React intent client). - Support-ticket ingestion (Zendesk, Intercom, Plain) linked to accounts. - Data-warehouse connections: scheduled read-only queries from Snowflake, BigQuery, Databricks, Redshift and ClickHouse. - Email-forwarded report ingestion into a dedicated agent inbox (forwarding rule or add as recipient). - CSV, spreadsheet and file import, including attaching a file in agent chat. - First-party website tracking script and embeddable forms capturing visitor identifiers and page/form interactions. - Document-open tracking (proposal opened N times, which sections). - Automatically permissioned multiplayer ingestion so each teammate's data is shared per policy. - Read-time freshness: records older than a threshold are re-checked when read; email re-verified before send. - Knowledge connectors (Drive, SharePoint, Confluence, Notion, Box, Dropbox, Egnyte, Highspot, Seismic, sitemap crawl). - Permission-aware indexing that respects source ACLs; Slack read in real time and not stored. - RFQ and order-document parsing from email bodies, PDF attachments and call transcripts into structured line items. - Bot-free transcript import from Zoom, Google Meet and Microsoft Teams. --- Source: https://www.closefinch.com/product/research # Research and signals Account and person research with cited sources, enrichment that fills only empty fields, and signals that say why an account is worth a look now. Facts quote their source; what the agent infers is labelled as inference. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Research and enrichment - Waterfall enrichment for email, phone and firmographics: free steps (CRM, website, email patterns) first, then paid providers charged only on a hit, with per-field provenance. *(Paid providers need your own key; the free steps run without one.)* - Search for companies and people beyond your CRM through People Data Labs, previewed with sources and CRM matches before a person imports them. *(Needs your own People Data Labs key; without one, the same filters run over your CRM.)* - Email verification before send (valid, invalid, catch-all, risky or unknown) from in-house checks, MillionVerifier and your own bounce history. *(Catch-all detection needs a MillionVerifier key.)* - Phone-number validity states (valid, invalid, unknown) with manual add. *(Carrier lookups need your own Telnyx account.)* - Picks the phone number most likely to connect, with the reason, and flags dead numbers. *(Off until set up.)* - Cost-disciplined enrichment: fill only missing fields from an approved provider under a cost and source policy; paid lookups need approval; no re-billing of fields already enriched. - Per-prospect and per-account deep research with citations and custom research prompts. *(Web search needs your own search provider key; without one, research uses the company's own site and news.)* - Natural-language research agent that plans sources and returns structured findings with evidence, ranked by fit and recency (minutes to hours), with a visible trace. - Live web re-check of facts just before a message is sent. - Public-records research: SEC filings and UK Companies House officers and filings, as dated findings with sources. - Niche-source lists from online stores, job boards, GitHub and a plugin marketplace, with quoted values and a fit verdict per row. *(Off until set up. Google Maps and other marketplaces use your own Apify account.)* - Web scraping as an agent tool on the built-in reader, honouring robots.txt and your allow and block lists. *(Off until set up. Uses your own Apify account.)* - Department headcount by function, tracked per quarter. *(Estimated from People Data Labs when keyed; otherwise counted from your CRM.)* - Living account plan or dossier: priorities, pain, people, tools, risks, next move. - Buying-committee map with each person's role and stance. - Business understanding from a pasted website URL (offer, ICP and proof points), refreshed as the business changes. - Brand and sender voice learned from website, emails, scripts and recorded calls. - Case-study miner: reads closed-won calls and notes for quantified outcomes and files proof points under business logic. - Competitive intelligence and battlecards kept current. - Research output that labels facts, inferences and unknowns separately. - Customer discovery-interview synthesis. - Individual briefs on stakeholders (background, past deals, how they review contracts) generated on request from Slack. ## Signals - Custom signals defined in plain English, for CRM changes or public events, with a category (intent, hiring, growth, events, engagement, champions or risk), a cadence and a strength score. - Signal monitor list with feed and all views, activity counts per monitor, and schedules in your time zone. - Company-level website visitor identification from network data and signed email links; raw IP addresses are never stored. - First-party page intent: visits to pages you tag (such as pricing and security), return visits, time on page, and several people from one account visiting within an hour. - Product signups and PLG usage signals, including a new user from an existing customer opening an expansion opportunity. - Job-change and champion-move tracking, including title change and email bouncing on known people. - Funding-round signals from SEC Form D filings and news, alerting the account owner as each round is found. - Hiring signals from an account's own careers pages and public job boards, with time filters and posting language matched to your pains. - Leadership and executive hires, new CFO, promotions. - Tech-stack change (from job posts, DNS and site) and competitor churn (a competitor logo leaving a customer page; a renewal that didn't happen). - Headcount growth by function, flagging teams that outgrow the rest of the company. *(Uses People Data Labs when keyed; otherwise your CRM's counts, if you opt in.)* - New office or location detected from postings and filings. - Social engagement signals: likers and commenters on your posts, competitor and influencer post engagers, viral posts, keyword posts. *(Off by default. Needs your own Unipile account.)* - LinkedIn profile viewers you import by hand, as signals. *(Off by default.)* - Recent-poster targeting: prospects who posted about their work in the last 7 days. *(Off by default. Needs your own Unipile account.)* - Explicit-need social listening on LinkedIn, X and Reddit that never infers intent from likes or views alone. *(Off by default. Needs your own Unipile account; X and Reddit need your own API keys.)* - Conference, event and webinar attendee or speaker lists matched to accounts and ICP roles. - Podcast-guest detection with the quote that opens the conversation. - Review-site signals from reviews you paste or upload as CSV: named pains, roles and company size, with quotes. - Third-party intent data from Bombora, G2 and 6sense. *(Off by default. Needs your own Bombora, G2 or 6sense account.)* - Partner, agency or investor mentions of an account. - Developer-community signals: activity on your GitHub repositories, mapped to accounts. - Email-engagement signals (repeated opens, link clicks) as triggers. *(An owner turns tracking on, and each sequence opts in.)* - Former- and past-customer detection (read-only Stripe key; past customers and lost deals re-located). - Shared-background signals (alma mater, same city, mutual history). - CRM- and deal-derived signals (stage changes, stalled deals, reply events, engagement drops) as agent triggers. - Real-time signal alerts to a Slack channel or owner DM with drafted action attached. - Signal stacking: open a play only when two or more signals coincide. - Signal analytics: which signals convert to meetings and revenue. - Dated why-now evidence attached to each account and prospect. - Customer spend trends: who is spending more and who is spending less. --- Source: https://www.closefinch.com/product/outbound # Prospecting and outbound ICP rules, lists and scoring, then sequences that draft each message from research, and an inbox for the replies. Do-not-contact rules, sending limits and deliverability checks apply to every send. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Prospecting - Natural-language people and company search over your CRM that compiles to removable filters, with saved matches. - ICP definition with may-contact and must-not-contact rules, and automatic exclusion of customers, partners and accounts with open deals. - Agent-proposed plays or ICPs generated by reading your website and profile, with targeting rules that persist per play. - Lookalike discovery from closed-won accounts, triggered on close, deduplicated against the CRM. *(Companies outside your CRM need an approved People Data Labs key.)* - AI qualification per account: custom fit questions answered pass / fail / unsure per row. - Account scoring and tiering by fit, signals and conversion likelihood, with the reasons for each account. - Prioritization model trained on the customer's own closed-won and closed-lost history. - Spreadsheet-style list workspace up to 10,000 rows: custom enrichment columns, CSV match-and-enrich, and AI cells only with a verified quote. - Agent-staged segments and audiences grouped by territory, segment or motion, with owner notification. - Day-one TAM grounded in ICP, existing customers and accounts found in your email history, auto-maintained. *(Market size needs an approved People Data Labs key.)* - Warm-path discovery: intros drafted for whichever teammate knows the account, and your team's network as a lead source. - Champion re-activation: find where past customers, lost-deal contacts and engaged buyers work now, verify, and alert. - Persona matching: choose the titles you actually sell to inside target or lookalike companies. - Lead-magnet manager: comment-a-keyword posts, deliver the resource, qualify and save genuine responders. *(Off by default. Needs your own Unipile account.)* - Waitlist ranking and design-partner recruiting that stops when the goal number of partners is active. - Local-business lead lists from OpenStreetMap or a CSV. *(Off until set up. Google Maps uses your own Apify account.)* - Daily ranked work list per rep with a stated reason and suggested action for every lead or account. - Event plays: invite leads near you to a dinner, coffee or conference meetup you host, with RSVPs and capacity. ## Sequences and outreach - Multi-step email sequences with step delays (typically 2–3 days), exit conditions and reply detection. - LinkedIn connection requests with a note, follow-up messages and InMail inside sequences, alongside email if you choose. *(Off by default. Needs your own Unipile account. Lanes are set through the API for now.)* - LinkedIn engagement steps: liking posts for everyone on a list, comments and reactions, connection refresh, and a follow-up when a request is accepted. *(Off by default. Needs your own Unipile account. Comment approvals and auto-like plays are through the API for now.)* - Rich media in messages: AI LinkedIn voice notes, AI avatar video from a photo, website screenshots, memes, personalized video. *(Off by default.)* - Messages written for each person from research, starting from why them and why now. - Commonality snippets: specific lines built from what you share with the prospect (alumni, named lookalike customers, shared conference), and agent outputs as template variables. - Sender-voice cloning and saved sender personas. - Prospect-style tone mirroring (calibrate to how the recipient writes). - Tone and CTA locks that the agent cannot override. - Drafting and replies in the recipient's language. - Visual play or journey graph with trigger, A/B, if/else, webhook, assign-owner and lookalike nodes, copy/paste and undo. - Adaptive sequences where the agent chooses channel, timing and format per lead and decides who to enroll and when. - Multi-sender enrollment and sender rotation across many domains and accounts. *(Off by default. Needs your own mailbox or Smartlead accounts.)* - Per-recipient-timezone send windows, blackout dates and scheduled pauses. - A/B and multivariate tests on messages, subject lines, sequences and audiences. - Reusable sequence rulesets (exclusions, eligibility, behavior) applied across sequences. - Send from the record owner's mailbox, with admin-managed signatures and booking links, and delegated sending. - Manual task steps (call, LinkedIn, email) mixed into agent sequences, with a daily task list for reps. - Push staged audiences into external sequencers (Instantly, HeyReach, Outreach, lemlist, Reply.io, Smartlead) as draft campaigns that launch only after approval. *(Off by default. Needs your own account with the sequencer.)* - Rule-based sequence pauses and bulk stop. - Paid-ads audience sync and retargeting of engaged prospects. *(Off until an owner connects your own ad accounts.)* - Physical gift and direct-mail plays through Sendoso. *(Off until an owner connects your own Sendoso account.)* - Autopilot campaigns that keep sourcing and enrolling new prospects after launch, with per-campaign refills. *(Off by default. Every email is a draft a person approves.)* - Event and webinar follow-up: match attendees to accounts and ICP roles, draft per person, hold for owner; registration and no-show plays. - Reactivation of closed-lost and long-dormant contacts, grounded in real relationship context, with approval required. - Snooze-aware follow-up: understands replies like not now or reach back in Q2, and schedules a re-engagement that pauses and resumes itself. - Open-quote follow-up cadence until the buyer purchases or opts out. ## Deliverability - Bring your own Gmail or Outlook mailbox via OAuth; teammates connect senders by invitation link. - Domain purchase (including lookalike domains) and Google/Outlook mailbox provisioning separate from the primary domain. *(Off by default. Needs your own Google Workspace and domain registrar accounts.)* - Continuous inbox warmup. *(Off by default. Needs your own Smartlead or Instantly account.)* - Inbox-placement and domain-health monitoring per mailbox and domain, with send pacing against mailbox health. *(Off by default. Seed tests need a placement-testing vendor; Google Postmaster data needs Google's approval.)* - Mailbox utilization and email-backlog dashboard. *(Off by default.)* - Pre-warmed inbox marketplace and migration of eligible domains to Gmail. *(Off by default. Needs your own Smartlead or Instantly account.)* - Sender caps per organization, and LinkedIn safety limits with a ramp-up. *(Off by default. LinkedIn steps need your own Unipile account.)* - Bounce prevention and automatic bounce handling. - Agent mailboxes on AgentMail: an agent sends and receives from its own address, through the same approvals, suppression and limits as people's mail, and replies come back through the same inbox. *(Needs your own AgentMail account.)* - Automated SPF, DKIM and DMARC setup and RFC 8058 one-click List-Unsubscribe headers on every outbound message. ## Replies - Unified inbox with research and history beside each thread, needs-reply and interested views, and color tags. *(LinkedIn messages join when Unipile is turned on.)* - Multi-label reply classification (positive, referral, objection, unsubscribe, not interested, later) where each label drives an action (end, pause, CRM write, alert, escalate, draft). - Out-of-office detection with auto-pause, auto-resume on the parsed return date, and exclusion from reply metrics. - AI-drafted replies grounded in business info and saved resources, approved by a human. - Autonomous replies that handle common objections, answer FAQs and recognize rejections. - Auto-reply to positive or eligible replies with a booking link. - Reply monitor that reads replies across sequencers and inboxes on a short schedule and messages the owner about the most urgent ones. - Escalation rules that hand complex or sensitive threads to a human. - Inbox-triage agent: each new email is triaged and a draft reply prepared. *(Gmail drafts and labels are off by default.)* - Inbound email routing by account ownership or round-robin. - Prompt-injection screening and untrusted-content isolation for inbound email and web pages before an agent acts on them. ## Nurture - Branded social post drafting and publishing to LinkedIn, X, Instagram and Reddit, with review and discover modes. - Flag commenters on your posts who sound like real buyers. *(Off by default.)* - Turn meeting moments into posts after consent and confidentiality checks. - SEO, local Google visibility and generative-engine (AI-answer) optimization content. *(Needs your own Serper key.)* - Agent-managed website presence improvements. *(Off until you connect WordPress.)* - Competitor and trend monitoring turned into a plan of launches, promotions, partnerships and local events. - Promotion launch with sales tracking. - Content-creator and partnership sourcing. - Nurture of leads and existing customers, drafted for approval. - Ads and analytics integrations (LinkedIn Ads, Reddit Ads, Google Ads, Analytics, Search Console) as agent tools. --- Source: https://www.closefinch.com/product/inbound # Inbound and scheduling A chat agent for your website, forms, lead qualification and routing, and a scheduler that books meetings on your team's calendars. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Website chat, forms and routing - Embeddable chat widget via JS snippet (theming, quick replies, CTA config, popup auto-open, teaser bubble). - Proactive chat opener for returning visitors, personalized to their signals. *(Off by default.)* - Engagement rules: who to engage, accounts to monitor without engaging, per-prospect instructions, daily message caps, and visitors who decline. - Pre-conversation profile from CRM history, prior visits, live intent and pages viewed, with cross-visit memory. - Matches a visitor to their CRM state (customer, open deal, closed-lost, owner) and shows the owning rep. - Your own qualifying questions plus ICP match, giving a numeric score and an agent-qualified lead. - Calendar gated on qualification (Book Meeting visible, but only high-propensity ICP buyers can book). - Answers from your knowledge base with source citations, a plain I don't know when it doesn't know, and a handoff to a person. - Private-context documents used for reasoning but never surfaced or linked to visitors. - Prices and limits copied from cited sources, never computed by the model. - Push-to-talk voice conversation on your website. *(Off until speech-to-text is set up and your site turns voice on.)* - Text and voice in one conversation, without asking the same questions again. - Modular on-page blocks (demos, videos, CTAs) rendered outside the chat. - Dynamic offers per buyer and per buying-committee member, timed to moments. - Agent-delivered first-call pitch, slide deck or site walkthrough. - Interactive product demos inside the chat and guided live product tours. - Greeting in the visitor's preferred language on arrival. - Live monitor of active visitors and conversations for reps. - Real-time high-intent alerts to Slack (pricing question, demo request, key-page visit), so a person can step in by chat, email or call. - Form builder and smart forms that display the scheduler instantly; agent-generated public forms. - Lead routing: round-robin within queues skipping out-of-office reps, territory by region/country/segment/vertical, account-ownership matching, branch routing. - Per-decision routing audit log showing which rule decided. - SDR→AE handoff with meeting credit attribution. - Inbound routing that grades every form fill or chat against your ICP and routes it with the reason. - Partner-referral intake from email and forms: create the account, link the partner, notify both owners. - Enterprise demo-request prep from your CRM: the likely committee, ownership and open deals, an agenda and a demo path. - Off-hours coverage that returns leads to SDR queues in business hours. - Lead intake from Facebook Lead Ads, Google Forms, Sheets, Typeform, CSV and portal leads. *(Off until set up.)* - Non-fit deflection and caller screening. - Malicious or sensitive conversation flagging and moderation that can end a conversation. - Community-channel agents (Slack communities, Discord, GitHub Issues and Discussions). *(Off until set up.)* - In-product guide agent (setup walkthroughs, activation and upgrade nudges). - Paid-ads landing-page conversion and ABM page personalization. - Two-way Slack conversation between rep and website agent, with live human takeover and transcript. ## Scheduling - Book directly on rep calendars after checking availability across multiple internal calendars. - Booking links via Calendly, Cal.com or Chili Piper inserted in replies and messages. - Native scheduler: booking pages, routing queues, group events, round-robin, time zones, a scheduler step in workflows, and agents booking for the team. - Mid-call voice booking with the invite or link sent by SMS or email before hang-up. *(Needs your own Telnyx account.)* - Booking onto field-service dispatch boards (ServiceTitan, Jobber, Housecall Pro). *(Needs your own ServiceTitan, Jobber or Housecall Pro account; automatic booking is off by default.)* - Appointment reminders and no-show rescheduling. *(SMS reminders need your own Telnyx account.)* - Meeting attribution: bookings matched to emailed contacts and counted per campaign. - Tentative hold on a calendar slot while the invite waits for approval. - Automatic follow-up scheduling after a send, with callbacks queued. --- Source: https://www.closefinch.com/product/meetings # Meetings Transcripts from Zoom, Google Meet and Microsoft Teams or from the recorder bot, notes and follow-ups after each call, and prompts during calls for people who opt in. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Meetings - Native recorder bot that auto-joins Zoom, Meet, Teams and Webex from the connected calendar. - Recording policy controls: record external meetings only or manual invite, custom bot name and image, waiting-room timeout, internal meetings tracked only with an outside attendee. - Live transcript with a running panel that turns promises into follow-ups, citing timestamps. - Real-time in-call assist: detect questions and objections, prompt the seller, surface live battlecards. *(Each person opts in.)* - AI ride-along participant that joins Zoom, Teams and Meet calls and drafts answers to technical questions for the seller. - Post-meeting summary, action items converted to tasks, drafted follow-up email and CRM update. - Meeting summary emails to your team's attendees, switchable in notification settings; outside attendees get the follow-up you approve. - Pre-meeting brief generated at a configured lead time: deal status, who's in the room (role, champion or new), history, risks, proof points, suggested questions, recommended next step. - Notification for meetings whose follow-ups never became tasks. - Add a transcript by hand, or attach a recording by link. - Meeting clips and automatic video snippets of key moments, plus a video library. - Topic tags, sentiment scoring and post-call analyzers that extract structured data and dispositions. - Per-call CRM field autofill for account, opportunity and contact fields, including methodology fields. - Batch extraction over existing records, and re-analysis of call history when an event fires. - Voice-of-customer, product-feedback, closed-lost-reason and churn-reason extraction feeding workflows. - Objection, competitor-mention and product-gap intelligence aggregated across calls. - Collateral and proposal generation after a meeting. - QBR preparation and solutions-engineer support (demo run-of-show, objection handling, fallback path). - Executive-assistant agent owning meeting prep, reminders, notes, follow-up and commitment checks. --- Source: https://www.closefinch.com/product/deals # Deals and forecasting Deal risk, next steps, coaching scorecards and forecasts, built from what buyers said on calls and in email. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Deal intelligence and coaching - Deal-risk flags: going cold relative to the expected pace for the stage, champion gone quiet, inactive contact, job-change risk, unanswered prospect email, ghosting, stalled next steps, missing decision maker, weak urgency. - Configurable risk thresholds per stage; a flag clears itself when the buyer replies. - Stalled-deal recovery: find deals with no activity in N days and draft a recovery note from the last call and business logic. - Infer and move deal stage from conversation signals (confirmed next steps, urgency, silence, exit criteria) under user-defined rules. - Record suggestions (a contact who keeps appearing isn't on the deal; a new lead matches an existing contact). - Methodology scoring: MEDDPICC evaluation per deal with a dedicated reasoning pass; 3 WHYs; methodology field autopilot. - Single-threading audit: missing roles, dormant champions, weak executive access, owner concentration. - One next step per deal, learned from what won deals did at the same stage. - Close plan and mutual action plan generation. - Value selling: business value assessment, value hypotheses and value pyramids. - Call coaching scorecards by stage (discovery, objection handling, next steps), AI coaching score, team performance and competency reviews. - AI roleplay practice for reps, in chat. - Win/loss analysis and deal diagnosis explaining why deals slip. - Auto-created Slack deal rooms and account rooms that route key moments, risks and updates to sales, CS and leadership. - Pipeline-review agent on a schedule per AE, and the same review on request in chat. - Sales-play recommendations and enablement-content curation. ## Forecasting - Forecast roll-ups with manager overrides. - Evidence-based commit test. Each commit must meet six conditions in the buyer's own words: confirmed intent, engaged economic buyer, agreed price and scope, dated close plan, procurement or legal started, close date in period. Produces two independent numbers (category roll-up with slippage haircut; stage-weighted from own win rates), conservative/expected/optimistic, flags dates pushed twice, and tracks accuracy against actuals. - Forecast submissions (rep and manager roll-up calls), pipeline snapshots and deal change history. - Scenario forecasting from micro and macro trends. - AI forecast rebuilt from call and deal evidence. - Weekly pipeline digest per AE posted before stand-up, plus CRO and executive briefs. - Pipeline coverage, slipped deals, stage velocity, win rates, and ARR split into new business and expansion. - Territory management and assignment by vertical and geography. - Quota setting and capacity planning. --- Source: https://www.closefinch.com/product/voice # Voice, SMS and messaging AI phone agents for inbound and outbound calls, two-way SMS, and WhatsApp, Telegram, Instagram and iMessage channels. Calls and texts run on your own Telnyx account, and outbound AI calling is off by default. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Voice and messaging - Inbound AI call answering at any hour, after hours and for overflow. *(Needs your own Telnyx account.)* - Speed-to-lead call within seconds of a form fill. *(Off by default. Needs your own Telnyx account.)* - Batch and broadcast outbound voice campaigns with scheduling, retry logic and goal tracking. *(Off by default. Needs your own Telnyx account.)* - CRM-event-triggered call workflows (deal-stage change, missed appointment, stalled quote). *(Off by default. Needs your own Telnyx account.)* - Voice persona: tone presets, a stock voice catalog and your own speech key. *(Premium voices need your own speech-provider key.)* - Mid-call channel switch from voice to SMS. *(Needs your own Telnyx account.)* - Buy numbers through Telnyx or bring your own over SIP, and run on your existing numbers. *(Needs your own Telnyx account.)* - Branded caller ID, spam-label remediation, number rotation and reputation monitoring. *(Off by default. Needs your own Telnyx account.)* - Call-screening assist that responds to carrier and iOS screening prompts. *(Off by default. Needs your own Telnyx account.)* - Answering-machine detection with a personalized TTS voicemail drop in the agent's voice; voicemail sequencer. *(Off by default. Needs your own Telnyx account.)* - Parallel dialing that connects the first live answer and cancels the rest (5–9 lines). *(Off by default. Needs your own Telnyx account.)* - Human power dialer and click-to-call with research context and AI call guide. *(Needs your own Telnyx account.)* - Phone-tree navigation to reach a human; hosted IVR. *(Off by default. Needs your own Telnyx account.)* - Live warm transfer to an available human with full context. *(Needs your own Telnyx account.)* - Calling in the US and Canada, and in other countries an owner turns on, with local-number country matching. *(Needs your own Telnyx account.)* - Deposit or payment collection on a call, with a confirmation text. *(Needs your own Telnyx and Stripe accounts.)* - Product-catalog reasoning on calls (SKUs, pricing tiers, regional rules) and recommendations from purchase history. *(Needs your own Telnyx account.)* - Mid-call tool calls to external APIs and a knowledge base. *(Needs your own Telnyx account.)* - Real-time voice conversations that handle interruptions and topic changes. *(Needs your own Telnyx account.)* - Emergency triage and dispatch with an escalation ladder and repeating notifications until a technician confirms. *(Needs your own Telnyx account.)* - Two-way SMS agent. *(Needs your own Telnyx account.)* - WhatsApp agent (WhatsApp Business). *(Needs your own WhatsApp Business account.)* - iMessage channel. *(Off by default. Needs your own Sendblue account.)* - Telegram and Instagram DM channels. *(Needs your own Telegram bot or Instagram account.)* - SMS alerts to the user when a prospect replies. *(Needs your own Telnyx account.)* - Unified human-plus-AI agent desktop. *(Needs your own Telnyx account.)* - Virtual sales floor with leaderboards and live listen. *(Needs your own Telnyx account.)* - Call recording, transcript, collected data and disposition codes synced to CRM, field-service or ERP systems. *(Needs your own Telnyx account.)* --- Source: https://www.closefinch.com/product/agents # Agents Agents built from templates or from a goal you describe. They run on a schedule, on an event or on demand, and show every step they took. Their writes go through the policy engine before they happen. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Agents - Agent and playbook template library with categories and role filters (Founder, Sales, Marketing, RevOps, CS), curated and community sections, one-click share and reuse. - Standard template anatomy: goal, numbered steps, who it's for, inputs, outputs, integrations with permission scopes, approvals, when it runs, what it replaces, estimated savings and FAQ. - Describe-a-goal agent builder that plans and runs multi-step work with visible step status. - Trigger types: schedule (cron), event (CRM change, deal closed-won, meeting recorded, new email), inbound (form, demo request, identified visitor), webhook, manual or on request. - Overnight scheduled skills that prepare leads, drafts and content into a morning queue. - Durable execution: agents do a bounded unit of work per run while the platform owns long-lived state (resumable runs, retries, follow-up cadence, full run history, replay, alerts). - Run trace where every step is a tool call tagged read, reason or write, with a note saying what ran. - Per-integration permission scopes labeled Reads / Writes new / Modifies existing, with optional assisted access, chosen before deploy. - Tool-level gates: require approval before a specific tool runs, disable tools, restrict tools to admins. - Approval queue for agent-generated actions with pending and completed counts and one-click approve, edit or reject. - Per-contact whole-sequence approval with inline per-step edits, available and kept in sync across the app (J/K keyboard review), a chat card and Slack buttons with an edit modal. - Approval expiry (default 10 days, configurable 1–90) with explicit status codes. - Autonomy levels from approve-every-message through per-contact approval and explicit per-sequence auto-send to full autopilot within guardrails. - Staged writes. Cascading actions summarize first and wait for confirmation. Merges, deletions, schema changes, owner changes, lifecycle updates and paid enrichment stay blocked until approved. External API/MCP callers cannot bypass owner approval. - Uncertainty escalation: the agent flags low confidence and waits, irreversible actions always confirm, and a queue holds tasks that need a person's decision. - Guardrails enforced in the send layer regardless of model output (suppression, do-not-contact, existing-customer and open-deal checks, deterministic controls). - Learning from human edits: an edited first message is saved as a marked example; Slack corrections retrain; skipped accounts and rewritten emails captured as team context. - Outcome-driven evals: each playbook version measured on replies, meetings, attendance, closed-won and sign-ups against version n−1; a regression benchmark gates every harness change. - Pre-launch simulations, auto-QA, live A/B of agent versions with winner promotion, conversation assessments and response scorecards. - Versioned, orchestrated agent deployment with version control. - Agents as managed staff: name, title, job description, org-chart position, a human manager, skill slots and run frequency. - Skills as portable Markdown files (SKILL.md) and installable skill packages or public skill libraries. - Deterministic code for predictable steps with LLM judgment reserved for hard steps; deterministic code sandbox; isolated per-agent cloud environments; agent writes and runs code. - Model routing by task cost and quality, a model your organization picks, and your own key if you bring one. Anthropic's models today. - Spend governance: daily budgets, cost confirmation before rewrites or data purchases, credit-balance stop that pauses rather than overcharges, live usage by category. - Agent lifecycle states (Draft, Starting, Working, Waiting on you, Done; Live, Ready) and run counters on agent cards. - Persistent agent conversations that can be handed to a teammate. - Project workspaces combining chat, context, audiences, files, people and canvases, shared with the team. - Business-logic knowledge base (positioning, proof points, voice, ICP, playbooks) that agents draw on. - Per-account persistent agents with memory, decisions and strategy exposed through an API. - Multi-agent orchestration: specialized agents for finding, writing and designing, a router that hands them work, and separate builder and worker agents. - Self-healing workers, and proposals for new angles or prospect pools when an agent's run quality dips. - Workflow builder with an agent step, rules engine, recipes and outbound HTTP/webhook blocks. - Explainability on every output: agents show their work, give the reason on each row, and cite the evidence behind each recommendation. - Recovery trail and dry-run for agent writes (verify every write; `--dry-run` and `--force` on destructive calls). - Auto-clear: an action or flag resolves itself once data or a reply shows the issue is fixed. --- Source: https://www.closefinch.com/product/quotes # Quotes and RFPs Quotes drafted from a deal description with your products, prices and approval rules, and questionnaire answers from a reviewed answer library. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Quotes - Quote drafted from a deal description or inbound request, applying your product, pricing and availability rules. - Agent pre-drafts the deal structure (commit, ramp, overage) from call transcripts and prior deals. - Rule-based approval routing (conditional, sequential or parallel multi-stakeholder), triggerable and answerable from Slack or an in-app approval queue. - Pricing-policy enforcement and discount guardrails with the human keeping final price control. - Usage, credit, token, outcome and seat pricing, alone or combined: tiers, ramps, monthly minimums, annual commits, overages, rollovers. - Setup agent that builds catalog, pricing, terms and approval rules from spreadsheets, order forms, CRM exports and recordings, tracing every value to its source. - Admin agent that checks rule changes for conflicts and generates test quotes; replay of past quotes against new pricing. - Per-number evidence trail: hovering any number shows the call, deal or rule behind it. - Approver context showing full pricing and approval history. - Quote output to order forms or slide decks, with plain-language change requests. - Renewals, amendments, expansions and cancellations in one composable quote; renewal quotes from live usage data. - Contract lifecycle for hybrid deals (commit treatments, credit rollovers). - Pricing intelligence from history: win/loss, margin trends, competitive positioning, how this buyer was priced before. - E-signature send and signature detection that moves the deal to Closed Won and triggers onboarding handoff. - Signed quotes pushed to NetSuite, Metronome or Stripe as customers, orders, contracts and subscriptions, each write approved by an owner and read back to confirm; invoice, payment and contract status read back onto the deal; the signed order form attached to the HubSpot or Salesforce deal. *(Needs your own NetSuite, Metronome or Stripe account.)* - Open-quote pipeline tracking (count, value, age, follow-up state). ## RFPs and questionnaires - Import Word and Excel questionnaires with AI detection of questions and sections; export back to the original format. - First-draft answers only from company-approved sources, with exact citations, a confidence score and a step-by-step generation trace. - Live-synced knowledge connectors, including compliance evidence (Vanta). *(Off until you connect Vanta.)* - Answer-library dedupe and stale-answer flagging. - Response workflow: owner, writer and reviewer roles, question-level comments and assignments, deadlines at questionnaire, section and question level, reminders, approval and sign-off. - Create an RFP project automatically from a CRM opportunity. - Slack quick-ask knowledge agent. - Team-workload and response-volume analytics, exposed via MCP. - Ad-hoc buyer questionnaire answering from governed company data through an MCP client. - PDF questionnaires answered in a separate answers file, and full library export. --- Source: https://www.closefinch.com/product/customers # Customer success Renewals, churn signals, onboarding handoffs, surveys and collections. Outreach is drafted for a person to approve. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Customer success - Renewal radar: watch renewals 120 days out, score risk from usage, replies and champion changes, brief the owner. - Date- and event-triggered lifecycle outreach templates: renewal and warranty reminders, service and reorder reminders, and onboarding plays. - Churn signals (champion departures, headcount drops, support tone, sentiment) that open a retention play. - Customer health scorecards, customer grids, retention forecast boards, product-usage and support-ticket tracking per account. - White-space mapping, cross-sell and upsell detection, expansion signals. - Dormant or declining account detection from order and spend history, with a drafted re-engagement opener and win-back trigger. - Onboarding agent: 90-day plan to first value with milestone tracking. - Closed-won handoff: kickoff task, handoff note to an onboarding channel, handoff-notes skill. - NPS surveys drafted for approval, with alerts and feedback synced to the CRM. - Payment follow-up and collections from Stripe invoices, drafted for approval. --- Source: https://www.closefinch.com/product/analytics # Analytics Questions answered in plain English with citations, dashboards, campaign and team views, attribution and exports. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Analytics - Campaign analytics: pending approvals, reply rate, emails sent, progress, meetings and per-contact status for each campaign. *(Through the API for now.)* - Attribution from signal, campaign or touchpoint to meetings, opportunities and closed-won revenue, including UTM passthrough. - Team view: bookings, results and focus by rep, region and date; rep performance against plan. - Questions in plain English over CRM, transcripts, email and Slack, answered with a citation for each statement and how fresh each source is. - Natural-language dashboards and agent-written code that produces charts, views and cross-account analysis. - Scheduled dashboard emails to your team: days, local time and time zone, recipients; edit, pause and resume; saved queries rerun; changes since the last email. - Morning brief for each rep by email or Slack DM: calls to make, accounts gone quiet and what changed overnight. - Agent telemetry: run counts and status per agent, per-task cost, resolve rate, live usage by category. - Live dashboards and live call monitoring. - Dollar-valued risk and opportunity roll-ups. - CSV, API and bulk exports for analysis in Claude, Codex or a warehouse. --- Source: https://www.closefinch.com/product/security # Security, admin and compliance Roles and fine-grained access, an audit log of every action, two-step sign-in and passkeys, scoped API keys, and consent and opt-out records across email, SMS and calls. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Sign-in and keys - Two-step sign-in with an authenticator app and encrypted backup codes, required for owners. - Session list and sign-out everywhere. - Breached-password check. - Sign-up abuse controls and trial limits. - Per-org send ramp and reputation. - Detection rules engine over audit and security events. - Hash-chained audit log with hourly verification. - Platform-wide kill switches and org freeze. - Incident runbook, breach notification and drills. - Passkeys (WebAuthn) for sign-in and as the second step; required for operators, offered to owners and members, with TOTP kept as a fallback. - API keys expire by default (90 days, up to 365; no expiry only as an owner's explicit choice, recorded) with reminders to the creator 14 and 3 days before. - CLI sign-in through the OAuth device flow with a built-in first-party client, so nobody needs to paste a long-lived API key; keys stay for CI. - Per-key address allowlist (CIDR ranges): a request from elsewhere is refused and recorded as a security event. - Fine-grained API scopes per resource and action, with named bundles; sending always needs an explicit send scope. - Log each use of a connected-account credential: which connection, for which job or request, by which person or agent, kept with the audit log. - Each production process gets only the secrets it uses (the web app gets no encryption keys or system database password), checked at start-up. ## Admin and data controls - Roles (Owner, Member, Viewer) with owner-only settings for pipelines, notification policy, API keys and org integrations. - Granular access by data source, agent, user role and team, including which AI tools can see which data. - Single sign-on through OIDC, and SCIM provisioning. *(Connects to your own identity provider.)* - Audit log of every query and action, including MCP queries, mapping approvals and per-decision reviewer and time. - Notification policy: per-event, per-channel org defaults and force-off locks, with personal preferences. - Scoped API keys and partner OAuth apps. - Logical multi-tenancy with each organization's data isolated, and separate deployments per region. - AES-256-GCM encryption of stored credentials and secrets under a versioned keyring with rotation and re-wrap; OAuth tokens deleted on disconnect and revoked at the provider once no other connection uses the grant; write scopes asked only when sending is allowed. - Data deletion on disconnect (embeddings and caches removed) and bulk export windows. - No training on customer data, zero-retention model options and training-provider exclusion on the model router. - Several brands and workspaces in one account: up to 5 brands, and a workspace per team or market. - Agent-governance layer: agent identity, permissions, auditability, workflow inventory, decision history and incident reports. - Org deletion with a 30-day pending window: the org is suspended at once, every owner is emailed, an owner can undo it, and it's purged when the window ends. ## Consent and compliance - Global do-not-contact list: named exclusions, CSV upload, matching across identifiers, rejected or deselected people auto-added. - Unsubscribe detection in replies with automatic suppression and preservation of unsubscribe rules across agents. - SMS keyword opt-out (STOP, UNSUBSCRIBE, CANCEL, END, QUIT and more), universal or per channel, logged with identifier, channel, keyword and time; HELP gets an information reply. - Calling-window engine: calling hours in every time zone a contact may be in, federal holidays, per-state hours and daily caps, state-of-emergency blocks, and a queue for held calls. - Do-not-call scrubbing against your own FTC registry file and internal list, with flagged numbers masked. *(Gryphon is optional and off.)* - Required disclosures: AI, recording and business-identity lines at the start of every call, transfer, voicemail and recording-stopped announcements, and an optional AI sender name on email. - Recording consent flow: first-use disclosure screen, explicit consent before recording, manual-notes fallback. - Consent and opt-out status written back to the CRM; consent records available for audit. - Card, SSN and bank-account redaction in transcripts before storage and model reads. - Exportable compliance audit history and regulator-shaped governance reports. - Unified per-contact consent and lawful-basis ledger spanning email, LinkedIn, SMS and voice. - Touch-frequency cap per contact across channels. - Known-relationship send gate: block or require approval for any outreach to a contact with no prior two-way interaction. ## Disclosures - EU AI Act disclosure. --- Source: https://www.closefinch.com/product/platform # Apps, API and MCP The web app, the Slack app and a mobile Today view, plus a REST API, a hosted MCP server, a CLI and webhooks for building on top. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Apps and developer surfaces - Web app: a home page with tasks, activity and metrics, plus records, agents, approvals and settings. - Slack app: alerts to channels and DMs, record link unfurls, two-way agent chat and @-mention Q&A, approval buttons, channel messages that trigger multi-step agent work with live progress lines. - Email notifications and digests (assignments, mentions, form submissions, daily suggestion list, meeting summaries). - Mobile Today view with meetings, tasks and approvals, including approving with a tap. *(Push notifications are off for now.)* - CLI built for agents: OAuth profiles, full CRUD per resource, raw API calls, bulk create, machine-readable `agent-context`, TTY tables vs. piped JSON, TOON output, `--jq`, dry-run and force; coding-agent installer. - Public REST API with OpenAPI, scoped keys, opaque prefixed IDs, cursor pagination, per-resource JSON Schema endpoints and published rate limits. - Webhooks (inbound triggers and outbound events) and Zapier/Make/n8n hooks. - Hosted MCP server with OAuth sign-in (clients can register themselves) or an API key, organization selection, scopes and gated tools. - MCP as a write and configuration surface (send email, quote from an LLM, push pricing config, replay regression tests). - MCP client: the agent consumes any remote MCP server or arbitrary API with keys stored as secrets. - WebMCP, so buyers' own AI assistants can query your site agent. *(WebMCP is a draft standard, in a Chrome origin trial.)* - Embeddable agents and generated apps (dashboards, work queues, public forms) inside CRM views or internal tools. - Email-the-agent interface (a forwarding inbox that ingests and acts). - Sidebar grouped by task: Today, Pipeline, Conversations, Prospect, Close, Agents, Coach (labelled, collapsible, at most two levels; sub-pages become tabs in their section); no page removed, and a test fails if any app route becomes unreachable. - Modules shown by setup: a module the org hasn't set up (voice, phone menus, payments, packs…) moves under More / Add a capability, one click away and still found by search. - Pinned and recent pages and records at the top of the sidebar, saved per person. - Settings as its own area with sections (Personal, Organization, Data, AI and agents, Channels, Phone and voice, Developers, Security) and a settings search. - Command palette (Cmd/Ctrl+K): jump to any page, find any record across all objects including custom ones, run actions (New deal, Ask…); recent items when empty, shortcuts shown beside commands, aware of the current page, only what the person's scopes allow. - App-wide keyboard shortcuts: ? help sheet, g-then-letter navigation, c to create, / to search, j/k through lists; the same on Mac and Windows, never fired while typing, and can be turned off. - On the free plan, every empty page links to the setup that fills it, plan limits say what stops and what still works before you reach them, and a short tour shows what you can do. ## For agents - Agent-created sandbox org that a person claims. - An agent can get started alone: a CLI command and an API call create a sandbox and return a working, limited test key, and API and MCP errors name the cause, the fix and a docs link, including where to get a key. ## Getting started - Guided first run. --- Source: https://www.closefinch.com/product/industries # Industry packs Packs for industrial distribution, equipment dealers and rental, consumer goods, home services and fundraising. Each adds its own data, triggers and agents. A note in italics after an item says when it's off by default, or needs your own account with another service. Every area: the [product overview](https://www.closefinch.com/product). ## Consumer goods - Distributor and retailer data ingestion (UNFI, KeHE, VIP depletions, Tesco, Sainsbury's Nectar360, dunnhumby) from emailed reports, downloads and EDI files. - Product, store and depot master data; a store that changes distributor keeps one record and its history. - Action types with definitions: unsold accounts, distribution gaps vs. core SKUs, voids, store out-of-stocks, phantom stock, DC reorders, low velocity, below last year, fill-rate short-ships. - False-churn prevention: check for distributor switches and short shipments before flagging an account. - Account-health buckets (Active, Slowing, Dormant, Lost) priced by what the account ordered before going quiet; 30/60/90-day unsold windows by channel, distributor or rep. - Dollar-ranked daily action list with drafted email to the responsible buyer, distributor rep or broker; approve-and-send; follow-up scheduled; auto-clear on resolution. - Category management: share vs. category and private label, velocity by retailer with movers flagged, promo lift vs. baseline, buyer-review deck. - DC reorder quantities in pallets from a run-rate window, target weeks of cover and open POs from the ERP. - Core-SKU coverage grid (store × SKU, by rep) with gaps ranked by size of prize. ## Equipment dealers and rental - Department-specific intake (parts inquiries, service requests, rental inquiries) with urgent requests routed to the right branch. - One view of what's open on an account: quotes, rentals, service tickets and receivables across systems. - Fleet utilization insights (earning vs. sitting) with add, remove or reposition recommendations across locations. - Asset and model-level profitability by make, model and class. - Fleet lifecycle analytics advising when to sell, based on maintenance trends, utilization and market values. - Lifecycle triggers: rental and lease renewals (90/60/30 days), warranty expiry, parts reorder, maintenance reminders, aftermarket upsell. *(Renewal email drafting is off by default.)* - Availability-aware rental booking and dispatch coordination (job routing, real-time dispatch). - Digital rental-return and periodic asset checks with damage documentation and condition tracking. - Field-rep mobile copilot with voice-recorded visit notes logged to the CRM. *(Voice notes need a speech-to-text key.)* ## Fundraising - Fundraising pack: a CSV round trip with YC Bookface's fundraising CRM, plus relevant investors and warm paths. *(Off until an owner turns it on.)* ## Home services - Home-services pack: estimate follow-up that waits while the customer thinks it over, missed-call recovery, and on-call dispatch at any hour with triage policies. *(Calls and texts need your own Telnyx account.)* - SMB vertical qualification logic (budget thresholds, insurance verification, address serviceability) with FSM and practice-management writeback. ## Industrial distribution - RFQ capture from email, web forms and call transcripts, with spec extraction. - Catalog spec matching (grade, size, alloy) and configured quoting. - Real-time inventory and availability checks against the ERP while quoting. *(Needs your ERP connected.)* - Netback pricing (net price after freight and other costs) informed by the buyer's price history. - Slow-moving inventory matched to past buyers of that grade and size. - Open-quote pipeline tracked with scheduled follow-ups until each quote is resolved. --- Source: https://www.closefinch.com/changelog # Changelog Release notes for Closefinch, newest first. API and MCP server versions, changes and deprecations are in the [developer changelog](https://www.closefinch.com/docs/changelog). ## Closefinch at launch *2026-11-08.* The first public release. A CRM that fills itself in from your team's email, calendars and meetings, and agents that ask before they act. ![A deals board for a demo team: deals in columns by stage, each with its amount, account and owner, and each column's total](https://www.closefinch.com/shots/deals-light.png) ![A demo account's page: its details on the left and, on the right, a timeline of the emails exchanged with its contacts](https://www.closefinch.com/shots/account-light.png) ![A demo team's meetings: each call's name, start and end, video link and owner](https://www.closefinch.com/shots/meetings-light.png) ![Part of a demo team's forecast: the roll-up by person, and the deals behind it with their amount, stage and forecast category](https://www.closefinch.com/shots/forecast-light.png) Closefinch captures a team's email, calendars, meetings and Slack into accounts, contacts and deals. A value it fills in links back to the email or call it came from, so you can check it. Keep your current CRM and sync with it, or use ours. Agents work from those records. They run on a schedule, on an event or when you ask, and show every step they took. Emails, texts and other actions outside the app pass the policy checks and, by default, wait for a person's approval. Website chat and phone agents answer live, within the limits you set. Your records and agents are also open to your own tools: a REST API, a hosted MCP server, a command line and webhooks. ### What's in it - **[CRM and data capture](https://www.closefinch.com/product/crm):** accounts, contacts, deals, leads, tasks and notes, with custom fields and objects, filled in from connected email, calendars, meetings and Slack. - **[Research and signals](https://www.closefinch.com/product/research):** account and person research with cited sources, and enrichment that fills only empty fields. - **[Prospecting and outbound](https://www.closefinch.com/product/outbound):** ideal customer profiles, lists and sequences, with do-not-contact rules and sending limits on every send. - **[Inbound and scheduling](https://www.closefinch.com/product/inbound):** a chat agent for your website, forms, lead routing and a meeting scheduler. - **[Meetings](https://www.closefinch.com/product/meetings):** transcripts from Zoom, Google Meet and Microsoft Teams or from the recorder bot, with notes and follow-ups after each call. - **[Deals and forecasting](https://www.closefinch.com/product/deals):** deal risk, next steps, coaching scorecards and forecasts, from what buyers said on calls and in email. - **[Voice, SMS and messaging](https://www.closefinch.com/product/voice):** AI phone agents and two-way SMS on your own Telnyx account. Outbound AI calling is off by default. - **[Agents](https://www.closefinch.com/product/agents):** agents from templates or from a goal you describe. - **[Quotes and RFPs](https://www.closefinch.com/product/quotes):** quotes drafted from a deal description with your products, prices and approval rules, and questionnaire answers from a reviewed answer library. - **[Customer success](https://www.closefinch.com/product/customers):** renewals, churn signals, onboarding handoffs and surveys, with outreach drafted for a person to approve. - **[Analytics](https://www.closefinch.com/product/analytics):** questions answered in plain English with citations, and dashboards. - **[Security, admin and compliance](https://www.closefinch.com/product/security):** roles, an audit log of every action, two-step sign-in and passkeys, and scoped API keys. - **[Apps, API and MCP](https://www.closefinch.com/product/platform):** the web app, the Slack app, a REST API, a hosted MCP server, a command line and webhooks. - **[Industry packs](https://www.closefinch.com/product/industries):** packs for industrial distribution, equipment dealers and rental, consumer goods, home services and fundraising. Everything, item by item, is on the [product pages](https://www.closefinch.com/product). API and MCP server versions are in the [developer changelog](https://www.closefinch.com/docs/changelog). --- Source: https://www.closefinch.com/pricing # Closefinch pricing Prices in USD, per month. Last updated 2026-11-05. Machine-readable: https://www.closefinch.com/pricing.json | Plan | Annual billing | Monthly billing | Seats | Recording hours | AI credits | Available | |---|---|---|---|---|---|---| | Free | $0 | $0 | 1–3 | 5 hours a month per workspace | 500 credits a month per workspace | Yes | | Team | $35 per seat | $44 per seat | 1+ | 10 hours a month per seat, pooled | 1,000 credits a month per seat, pooled | Not yet | | Pro | $79 per seat | $99 per seat | 1+ | 20 hours a month per seat, pooled | 2,500 credits a month per seat, pooled | Not yet | ## Free For a small team trying it on real work. - CRM: accounts, contacts, deals, leads, tasks, pipelines and custom fields - Email and calendar capture from Gmail or Microsoft 365 - Meeting transcripts imported from Zoom, Google Meet, Teams and notetakers - Meeting recorder bot for Zoom, Meet, Teams and Webex - Agents with approvals: nothing leaves without a person, by default - REST API, MCP server and CLI Available now. No card needed. Until billing launches, AI use has a daily trial cap instead of credits. ## Team For a sales team that wants the CRM to keep itself up to date. - Everything in Free, with no seat cap and more recorder hours - Two-way sync with Salesforce, HubSpot, Attio and Pipedrive; write-through to Close, Zoho, Affinity and DealCloud - Agent templates, schedules and event triggers - Slack app, knowledge connectors, deal risk and next steps Opens when billing launches. Until then, new workspaces start on Free. ## Pro For teams that run forecasts and coaching on the same data. - Everything in Team - Forecasting: evidence-based commit test, roll-ups, scenarios - Call coaching scorecards and roleplay practice - Slack deal rooms, win/loss analysis and dollar-valued risk roll-ups - More recording hours and AI credits Opens when billing launches. Until then, new workspaces start on Free. ## Compare plans ### Price and limits | | Free | Team | Pro | |---|---|---|---| | Billed annually | $0 | $35 per seat a month | $79 per seat a month | | Billed monthly | $0 | $44 per seat a month | $99 per seat a month | | Seats | Up to 3 | No limit | No limit | | Recorder hours | 5 hours a month per workspace | 10 hours a month per seat, pooled | 20 hours a month per seat, pooled | | AI credits | 500 credits a month per workspace | 1,000 credits a month per seat, pooled | 2,500 credits a month per seat, pooled | | Available today | Yes | Opens with billing | Opens with billing | ### What's included | | Free | Team | Pro | |---|---|---|---| | CRM: accounts, contacts, deals, leads, tasks, pipelines and custom fields | Yes | Yes | Yes | | Email and calendar capture from Gmail or Microsoft 365 | Yes | Yes | Yes | | Meeting transcripts imported from Zoom, Google Meet, Teams and notetakers | Yes | Yes | Yes | | Meeting recorder bot for Zoom, Meet, Teams and Webex | Yes | Yes | Yes | | Agents with approvals: nothing leaves without a person, by default | Yes | Yes | Yes | | REST API, MCP server and CLI | Yes | Yes | Yes | | Two-way sync with Salesforce, HubSpot, Attio and Pipedrive; write-through to Close, Zoho, Affinity and DealCloud | — | Yes | Yes | | Agent templates, schedules and event triggers | — | Yes | Yes | | Slack app, knowledge connectors, deal risk and next steps | — | Yes | Yes | | Forecasting: evidence-based commit test, roll-ups, scenarios | — | — | Yes | | Call coaching scorecards and roleplay practice | — | — | Yes | | Slack deal rooms, win/loss analysis and dollar-valued risk roll-ups | — | — | Yes | - Until billing launches, plans don't limit features or seats: a Free workspace can use everything listed here. The feature and seat limits above start with billing. - Free: Available now. No card needed. Until billing launches, AI use has a daily trial cap instead of credits. - Team: Opens when billing launches. Until then, new workspaces start on Free. - Pro: Opens when billing launches. Until then, new workspaces start on Free. ## Usage - **Bot-recorded meeting hour beyond the plan's hours:** $1.25 per hour. Only meetings the recorder bot joins use hours. Meetings whose platform transcript we import (Zoom, Meet, Teams) don't. - **AI credit pack:** $50 per 5,000 credits. Credits pay for the AI work agents and features do. Each run's trace shows what it used. ## Questions ### What is a seat? A person who can sign in to the workspace. Agents aren't seats. ### What uses recording hours? Only meetings the recorder bot joins. Transcripts imported from Zoom, Google Meet or Teams, uploads and notetaker imports don't use hours. Hours are pooled across the workspace and reset monthly. ### What are AI credits? Credits pay for model usage: summaries, drafts, field filling, answers and agent runs. One credit is $0.01 of usage at our list rates, which include our margin over the model providers' prices. Every run's trace shows its cost, and owners can set daily and monthly budgets; runs pause when one is reached. ### Is there a trial of the paid plans? Not yet. Paid plans open when billing launches. Until then every workspace starts on Free, with trial limits on sending, recording and AI use. ### Do prices include sales tax or VAT? Prices are in US dollars before tax. Paddle, our merchant of record, adds the sales tax or VAT that applies where you're billed, collects it, and shows it on each invoice. The billing page in the app shows totals with tax for your country before you pay. ### Do you train AI models on our data? No. We don't sell customer data or use it to train models. See the privacy policy and subprocessor list. --- Source: https://www.closefinch.com/integrations # Integrations Connect the tools your team already uses. Every connection is per person or per workspace, and nothing is sent out through one without passing the policy engine. ## Email and calendar Each teammate connects their own mailbox. Business email and meetings are filed to the right contacts and accounts; personal and marketing mail is left out. - **[Google (Gmail and Calendar)](https://www.closefinch.com/integrations/google):** Mail and meetings from Gmail and Google Calendar. - **[Google Calendar only](https://www.closefinch.com/integrations/google-calendar):** Meetings without access to Gmail. - **[Microsoft 365 (Outlook)](https://www.closefinch.com/integrations/microsoft):** Mail and meetings from Outlook and Outlook Calendar. - **[Outlook Calendar only](https://www.closefinch.com/integrations/microsoft-calendar):** Meetings without access to mail. ## CRMs Keep your current CRM. Closefinch syncs two ways, or writes through, and shows conflicts for a person to settle. Moving over instead? The migration tool imports from a connected CRM or files. - **[Salesforce](https://www.closefinch.com/integrations/salesforce):** Two-way sync of accounts, contacts, opportunities, leads, tasks and notes. Sandboxes too. - **[HubSpot](https://www.closefinch.com/integrations/hubspot):** Two-way sync of companies, contacts, deals, tasks and notes, with pipelines and owners. - **[Attio](https://www.closefinch.com/integrations/attio):** Two-way sync of companies, people, deals, tasks and notes. - **[Pipedrive](https://www.closefinch.com/integrations/pipedrive):** Two-way sync of organizations, persons, deals, leads, activities and notes. - **[Close](https://www.closefinch.com/integrations/close):** Write-through: leads, contacts, opportunities, tasks and notes are written to Close. - **[Zoho CRM](https://www.closefinch.com/integrations/zoho):** Write-through, in any Zoho data center. - **[Affinity](https://www.closefinch.com/integrations/affinity):** Write-through for organizations, people and opportunity lists. - **[DealCloud](https://www.closefinch.com/integrations/dealcloud):** Write-through for your site's company, contact and deal entry types. ## Meetings and notetakers Native transcripts come first, so no bot joins when the platform already made one. The recorder bot is there for the rest, with a notice and a stop link for everyone in the meeting. - **[Zoom](https://www.closefinch.com/integrations/zoom-meetings):** Imports Zoom's own transcripts of cloud-recorded meetings. - **[Google Meet](https://www.closefinch.com/integrations/google-meet):** Imports Meet's own transcripts. - **[Microsoft Teams](https://www.closefinch.com/integrations/microsoft-teams):** Imports Teams' own transcripts of meetings on your Outlook calendar. - **[Gong](https://www.closefinch.com/integrations/gong):** Calls with briefs, key points, next steps and transcripts. - **[Granola](https://www.closefinch.com/integrations/granola):** Meeting notes and transcripts. - **[Fathom](https://www.closefinch.com/integrations/fathom):** Summaries, action items and transcripts by webhook. - **[Fireflies.ai](https://www.closefinch.com/integrations/fireflies):** Transcripts and summaries. - **[Circleback](https://www.closefinch.com/integrations/circleback):** Notes, action items and transcripts by webhook. - **[Notion AI Meeting Notes](https://www.closefinch.com/integrations/notion-meetings):** Summaries, notes and transcripts. - **[Grain](https://www.closefinch.com/integrations/grain):** Summaries, action items and transcripts. - **[Chorus](https://www.closefinch.com/integrations/chorus):** Meetings and dialer calls with summaries and transcripts. - **[Plaud](https://www.closefinch.com/integrations/plaud):** In-person recordings through Plaud's Zapier trigger. - **[Any other notetaker](https://www.closefinch.com/integrations/notetaker-webhook):** A signed webhook for Avoma, tl;dv, Otter or any relay. ## Phone and dialers Calls are logged to the contact and account matched by phone number, with recordings and outcomes. - **[Aircall](https://www.closefinch.com/integrations/aircall):** Calls, outcomes, recordings and comments. - **[Dialpad](https://www.closefinch.com/integrations/dialpad):** Calls with recordings and dispositions. - **[RingCentral](https://www.closefinch.com/integrations/ringcentral):** Calls from telephony-session events. - **[Zoom Phone](https://www.closefinch.com/integrations/zoom-phone):** Calls, call logs, recordings and transcripts. - **[Orum](https://www.closefinch.com/integrations/orum):** Power-dialer calls with dispositions and notes. - **[Outreach](https://www.closefinch.com/integrations/outreach):** Calls made in Outreach's dialer. ## Slack Conversations from channels and Slack Connect land on the right account. Alerts, approvals and deal rooms work in Slack too. - **[Slack](https://www.closefinch.com/integrations/slack):** Channel and thread capture, Slack Connect, alerts, two-way agent chat and deal rooms. - **[Slack search](https://www.closefinch.com/integrations/slack-search):** The assistant searches Slack as you, with your own permissions, at question time. ## Knowledge Indexed for answers and agents, respecting each source's sharing so people only see what they could already open. - **[Google Drive](https://www.closefinch.com/integrations/google-drive):** Docs, Sheets, Slides, PDF and Office files. - **[SharePoint and OneDrive](https://www.closefinch.com/integrations/sharepoint):** Document libraries and OneDrive files. - **[Confluence](https://www.closefinch.com/integrations/confluence):** Confluence Cloud pages. - **[Notion](https://www.closefinch.com/integrations/notion):** Pages shared with the integration. - **[Dropbox](https://www.closefinch.com/integrations/dropbox):** Files with their sharing. - **[Box](https://www.closefinch.com/integrations/box):** Files with their sharing. - **[Egnyte](https://www.closefinch.com/integrations/egnyte):** Files under the paths you choose. - **[Highspot](https://www.closefinch.com/integrations/highspot):** Items in your spots. - **[Seismic](https://www.closefinch.com/integrations/seismic):** Library content. - **[Your website](https://www.closefinch.com/integrations/sitemap):** Pages from your sitemap. ## Product, billing and support signals What customers do in your product, what they pay and what they ask support, on the account's timeline. - **[Stripe](https://www.closefinch.com/integrations/stripe):** Customers, invoices, subscriptions and refunds. - **[PostHog](https://www.closefinch.com/integrations/posthog):** Product events linked by email and domain. - **[Segment](https://www.closefinch.com/integrations/segment):** Product events linked by email and domain. - **[Amplitude](https://www.closefinch.com/integrations/amplitude):** Product events linked by email and domain. - **[Mixpanel](https://www.closefinch.com/integrations/mixpanel):** Product events linked by email and domain. - **[Website and app events](https://www.closefinch.com/integrations/intent-js):** A first-party JavaScript snippet for page views and product events. - **[Zendesk](https://www.closefinch.com/integrations/zendesk):** Support tickets linked by requester. - **[Intercom](https://www.closefinch.com/integrations/intercom):** Conversations and tickets. - **[Plain](https://www.closefinch.com/integrations/plain):** Support threads. ## Marketing Your own website and ad and analytics accounts. Agents read reports; changes to your site or campaigns wait for a person's approval. - **[WordPress](https://www.closefinch.com/integrations/wordpress):** Approved page titles and descriptions, and drafts of approved articles, through an application password. - **[Google Ads, Analytics and Search Console](https://www.closefinch.com/integrations/google-marketing):** GAQL reports, GA4 reports and Search Console queries; pausing a campaign is approved by a person. - **[LinkedIn Ads](https://www.closefinch.com/integrations/linkedin-ads):** Campaign performance; pausing or enabling a campaign is approved by a person. - **[Reddit Ads](https://www.closefinch.com/integrations/reddit-ads):** Campaign performance; pausing or enabling a campaign is approved by a person. ## Data warehouses Read-only queries on a schedule bring warehouse rows onto accounts and contacts; records can be exported back. - **[Snowflake](https://www.closefinch.com/integrations/snowflake):** Scheduled read-only queries. - **[BigQuery](https://www.closefinch.com/integrations/bigquery):** Scheduled read-only queries. - **[Databricks](https://www.closefinch.com/integrations/databricks):** Scheduled read-only queries. - **[Amazon Redshift](https://www.closefinch.com/integrations/redshift):** Scheduled read-only queries. - **[ClickHouse](https://www.closefinch.com/integrations/clickhouse):** Scheduled read-only queries. ## Agent inbox Forward reports and emails to an agent's own address and they're filed and acted on. - **[Agent inbox](https://www.closefinch.com/integrations/agent-inbox):** A dedicated address for forwarded reports. ## For developers - **[REST API](https://www.closefinch.com/docs):** Every resource, OpenAPI 3.1, scoped keys, cursor pagination and published rate limits. - **[MCP server](https://www.closefinch.com/docs/mcp):** Connect Claude, ChatGPT, Cursor or any MCP client with OAuth or an API key. - **[CLI](https://www.closefinch.com/docs/cli):** The `ra` command line, built for scripts and coding agents. - **[Webhooks](https://www.closefinch.com/docs/webhooks):** Inbound triggers and outbound events (signed per Standard Webhooks), for Zapier, Make and n8n too. --- Source: https://www.closefinch.com/templates # Agent templates Built-in agents you can start from. Each one ships with an eval suite it must pass, runs on a schedule, an event or on demand, and asks before anything leaves. ## Pipeline - **[Act on signals](https://www.closefinch.com/templates/signal-follow-up):** When a signal fires on an account, turns its dated why-now evidence into one explained task for the owner. No evidence, no task. - **[Stalled deal nudge](https://www.closefinch.com/templates/stalled-deal-nudge):** Finds open deals that have gone quiet, decides one concrete next step for each, and leaves a follow-up task. - **[Stalled deal recovery](https://www.closefinch.com/templates/stalled-deal-recovery):** Every morning, finds deals with no activity for two weeks, works out why from the last call, and puts a recovery email in the owner's inbox to approve. - **[Pipeline review](https://www.closefinch.com/templates/pipeline-review):** Twice a week, reviews one AE's open deals (risks, methodology gaps, last touch, one next action) and sends an honest take. - **[Forecast manager](https://www.closefinch.com/templates/forecast-manager):** The day before the forecast call, tests every Commit deal against six conditions in the buyer's own words and shows the range with its formulas. ## Meetings - **[Meeting prep brief](https://www.closefinch.com/templates/meeting-prep-brief):** Before a call, reads the account's records, recent conversations and saved facts and writes a short cited brief. - **[Meeting assistant](https://www.closefinch.com/templates/meeting-assistant):** Prep a day before, a reminder 30 minutes before, notes, tasks and a drafted follow-up after, and a check three days later of which commitments were kept. - **[Demo prep for solutions engineers](https://www.closefinch.com/templates/se-demo-prep):** A timed run-of-show mapped to what the buyer asked to see, with a talk track, proof points and responses to the objections they raised. ## Customer success - **[Inbound receptionist](https://www.closefinch.com/templates/inbound-receptionist):** Answers your phone after hours, on holidays and when nobody picks up: says it's an AI first, answers from your knowledge base, takes messages and books meetings, then tells the right person what needs doing. - **[Emergency dispatcher](https://www.closefinch.com/templates/emergency-dispatcher):** Checks a reported emergency against your own list and pages your on-call ladder by text, call and app, repeating and escalating to the backup and a manager until a technician accepts; tells the customer honestly what happens next. - **[Account memory keeper](https://www.closefinch.com/templates/account-memory-keeper):** Saves durable facts from conversations (preferences, processes, risks) to the account, each with the quote it came from. - **[Won deal handoff](https://www.closefinch.com/templates/won-deal-handoff):** When a large deal is won, waits an hour, then opens the onboarding task for the team. - **[QBR prep](https://www.closefinch.com/templates/qbr-prep):** Prepares a quarterly business review: the evidence in the customer's words, the narrative, and a timed internal runbook. ## Reporting - **[Weekly pipeline summary](https://www.closefinch.com/templates/weekly-pipeline-summary):** Summarizes open pipeline and what's stalled into a headline, a short summary and the top risks. ## Data quality - **[Missing close dates](https://www.closefinch.com/templates/missing-close-dates):** Finds open deals without a close date and asks their owners to add one. No AI involved. ## Research - **[Account research brief](https://www.closefinch.com/templates/account-research-brief):** Researches an account's own site and recent news and sends a brief: every fact quotes its source; inferences and unknowns are labelled. - **[Person research brief](https://www.closefinch.com/templates/person-research-brief):** Researches a contact's public work through their company's site and the news that names them, with labelled, cited findings. - **[Account plan keeper](https://www.closefinch.com/templates/account-plan-keeper):** Keeps a living plan per account (priorities, pain, people, tools, risks, next move), every item sourced and dated, refreshed when signals fire; your edits stay. - **[Case-study miner](https://www.closefinch.com/templates/case-study-miner):** When a deal is won, files the customer's quantified results from its calls and notes as proof points with exact quotes, each waiting for your approval. - **[Competitive intel analyst](https://www.closefinch.com/templates/competitive-intel):** Monthly: logs what changed on competitors' pages, finds where they came up in your deals, and proposes cited battlecard edits for you to approve. - **[Business profile from your site](https://www.closefinch.com/templates/business-profile-from-site):** Reads your own website and proposes your positioning, ICP, proof points, competitors and plays, every claim quoting its page. Nothing is saved until a person accepts it. ## Prospecting - **[ICP list builder](https://www.closefinch.com/templates/icp-list-builder):** Each week, stages segments of the CRM accounts that fit your ICP by territory, leaving out customers, partners, open deals and must-not-contact accounts. - **[Fit qualifier](https://www.closefinch.com/templates/fit-qualifier):** Answers your plain-language fit questions for every account on a list: pass, fail or unsure, each with a reason and its source (a CRM value or a checked quote from the account's own site). Never a guess. - **[Design-partner recruiter](https://www.closefinch.com/templates/design-partner-recruiter):** Works your ranked waitlist until N design partners are active: tracks each candidate's stage and drafts the next emails for your approval, stopping the moment the goal is met. - **[Two-way SMS agent](https://www.closefinch.com/templates/two-way-sms-agent):** Answers texts to your number from your knowledge base, says it's an AI in its first text, and hands pricing, contracts, complaints or a request for a person to the thread's owner. Opt-outs, quiet hours and do-not-contact always apply. - **[Speed-to-lead caller](https://www.closefinch.com/templates/speed-to-lead-caller):** Calls a new lead back within seconds of their form, only if they agreed to calls on it and only inside their calling window; says it's an AI first, retries a few times, and books the meeting. Off until an owner turns it on. - **[Voice campaign caller](https://www.closefinch.com/templates/voice-campaign):** Adds people to a voice campaign (or lets a deal stage change, a missed meeting or a quiet quote do it): the AI calls them in their calling hours, retries, leaves a short voicemail in your agent's voice and stops at the goal. Only people with written consent are called. --- Source: https://www.closefinch.com/compare # Closefinch compared Published prices for tools people compare us with, from each vendor's own pricing page, checked 2026-11-05. Figures marked unverified weren't confirmed on the vendor's page. Product names belong to their owners. | Product | Kind | Plans | Free tier or trial | Source | |---|---|---|---|---| | [Attio](https://www.closefinch.com/compare/attio) | AI CRM | Free: $0, up to 3 seats; Plus: $35 per seat (annual), $44 (monthly); Pro: $79 per seat (annual), $99 (monthly); Enterprise: Custom | Free plan; 14-day Pro trial. | [attio.com/pricing](https://attio.com/pricing) | | [HubSpot Sales Hub](https://www.closefinch.com/compare/hubspot) | CRM suite | Free: $0, 2 users; Starter: From $7–20 per seat; Professional: From $90 per seat, plus $1,500 onboarding; Enterprise: From $150 per seat, plus $3,500 onboarding | Free tier. | [hubspot.com/pricing/sales](https://www.hubspot.com/pricing/sales) | | [Pipedrive](https://www.closefinch.com/compare/pipedrive) | sales CRM | Lite: $14 per seat (annual), $24 (monthly) (unverified); Ultimate: $79 per seat (annual), $99 (monthly) (unverified) | Trial; no free plan (unverified). | [pipedrive.com/en/pricing](https://www.pipedrive.com/en/pricing) | | [Close](https://www.closefinch.com/compare/close) | sales CRM | Solo: $9–19 per user; Essentials: $35–49 per user; Growth: $99–109 per user; Scale: $139–149 per user | 14-day trial. | [close.com/pricing](https://www.close.com/pricing) | | [Day.ai](https://www.closefinch.com/compare/day-ai) | AI CRM | Free: $0; Turbo: $24 per agent (annual); Professional: $60 per agent (annual); Executive: $200 per agent (annual); Super: $800 per agent (annual) | Free tier that includes capture. | [day.ai/pricing](https://day.ai/pricing) | | [Folk](https://www.closefinch.com/compare/folk) | CRM | Standard: $24 per member (annual), $30 (monthly); Premium: $48 per member (annual), $60 (monthly); Enterprise: From $80 per member | 2-week trial, no card. | [folk.app/pricing](https://www.folk.app/pricing) | | [Lightfield](https://www.closefinch.com/compare/lightfield) | AI CRM | Starter: $89 per user (monthly, up to 3 seats); Pro: $999 per workspace (annual); Growth: $2,999 per workspace (annual); Enterprise: Custom | 14-day trial; no free tier. | [lightfield.com/pricing](https://lightfield.com/pricing) | | [Gong](https://www.closefinch.com/compare/gong) | revenue intelligence platform | All plans: Quote only | Not published. | [gong.io/pricing](https://www.gong.io/pricing) | ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/attio # Closefinch and Attio Attio is an AI CRM. Its figures below come from its own pricing page, [attio.com/pricing](https://attio.com/pricing). Checked 2026-11-05. Product names belong to their owners. ## Attio's published pricing | Plan | Price | |---|---| | Free | $0, up to 3 seats | | Plus | $35 per seat (annual), $44 (monthly) | | Pro | $79 per seat (annual), $99 (monthly) | | Enterprise | Custom | - **Usage:** Credits per seat per month: 100 / 500 / 1,000 / 2,500 by plan. Packs from 5,000 for $70 to 50,000 for $475 (annual). - **Free tier or trial:** Free plan; 14-day Pro trial. ## Side by side | | Attio | Closefinch | |---|---|---| | Published prices | Yes, on attio.com/pricing | Yes, on this site, in pricing.md and in pricing.json | | Priced per | Seat | Seat (Free covers up to 3 seats) | | Free plan or trial | Free plan; 14-day Pro trial. | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | Credits per seat per month: 100 / 500 / 1,000 / 2,500 by plan. Packs from 5,000 for $70 to 50,000 for $475 (annual). | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together Closefinch syncs two ways with Attio (companies, people, deals, tasks, notes), so you can keep Attio as your system of record and add capture and agents on top. ## What differs - Our Team and Pro list prices are the same as Attio's Plus and Pro. - Our plans include meeting-recorder hours; extra hours are metered at $1.25. - Attio publishes its pricing as Markdown; so do we, plus pricing.json. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/hubspot # Closefinch and HubSpot Sales Hub HubSpot Sales Hub is a CRM suite. Its figures below come from its own pricing page, [hubspot.com/pricing/sales](https://www.hubspot.com/pricing/sales). Checked 2026-11-05. Product names belong to their owners. ## HubSpot Sales Hub's published pricing | Plan | Price | |---|---| | Free | $0, 2 users | | Starter | From $7–20 per seat | | Professional | From $90 per seat, plus $1,500 onboarding | | Enterprise | From $150 per seat, plus $3,500 onboarding | - **Usage:** Credits per month: 500 / 3,000 / 5,000 by plan; extra credits $0.01 each. - **Free tier or trial:** Free tier. ## Side by side | | HubSpot Sales Hub | Closefinch | |---|---|---| | Published prices | Yes, on hubspot.com/pricing/sales | Yes, on this site, in pricing.md and in pricing.json | | Priced per | Seat, plus a one-time onboarding fee on Professional and Enterprise | Seat (Free covers up to 3 seats) | | Free plan or trial | Free tier. | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | Credits per month: 500 / 3,000 / 5,000 by plan; extra credits $0.01 each. | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together Closefinch syncs two ways with HubSpot companies, contacts, deals, tasks and notes, with pipelines and owners. ## What differs - HubSpot sells sales, marketing and service products; Closefinch covers sales capture, CRM and agents. - We charge no onboarding fee. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/pipedrive # Closefinch and Pipedrive Pipedrive is a sales CRM. Pipedrive's pricing page didn't load when we checked, so these figures come from secondary sources and are unverified. Checked 2026-11-05. Product names belong to their owners. Vendor page: [pipedrive.com/en/pricing](https://www.pipedrive.com/en/pricing) ## Pipedrive's published pricing | Plan | Price | |---|---| | Lite | $14 per seat (annual), $24 (monthly) (unverified) | | Ultimate | $79 per seat (annual), $99 (monthly) (unverified) | - **Usage:** Not checked. - **Free tier or trial:** Trial; no free plan (unverified). ## Side by side | | Pipedrive | Closefinch | |---|---|---| | Published prices | Not confirmed: the page didn't load when we checked (unverified) | Yes, on this site, in pricing.md and in pricing.json | | Priced per | Seat (unverified) | Seat (Free covers up to 3 seats) | | Free plan or trial | Trial; no free plan (unverified). | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | Not checked. | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together Closefinch syncs two ways with Pipedrive (organizations, persons, deals, leads, activities, notes; pipelines and stages). ## What differs - We couldn't load Pipedrive's pricing page when we checked, so its figures here come from secondary sources. Check their page before deciding. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/close # Closefinch and Close Close is a sales CRM. Its figures below come from its own pricing page, [close.com/pricing](https://www.close.com/pricing). Checked 2026-11-05. Product names belong to their owners. ## Close's published pricing | Plan | Price | |---|---| | Solo | $9–19 per user | | Essentials | $35–49 per user | | Growth | $99–109 per user | | Scale | $139–149 per user | - **Usage:** AI credits 500 to 2,000 a month by plan; calls about $0.02 a minute. - **Free tier or trial:** 14-day trial. ## Side by side | | Close | Closefinch | |---|---|---| | Published prices | Yes, on close.com/pricing | Yes, on this site, in pricing.md and in pricing.json | | Priced per | User | Seat (Free covers up to 3 seats) | | Free plan or trial | 14-day trial. | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | AI credits 500 to 2,000 a month by plan; calls about $0.02 a minute. | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together Closefinch works with Close in write-through mode: leads, contacts, opportunities, tasks and notes are written to Close. ## What differs - Close includes calling; Closefinch logs calls from dialers you already use (Aircall, Dialpad, RingCentral, Zoom Phone, Orum, Outreach) rather than placing them. - Close publishes its pricing as Markdown; so do we, plus pricing.json. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/day-ai # Closefinch and Day.ai Day.ai is an AI CRM. Its figures below come from its own pricing page, [day.ai/pricing](https://day.ai/pricing). Checked 2026-11-05. Product names belong to their owners. ## Day.ai's published pricing | Plan | Price | |---|---| | Free | $0 | | Turbo | $24 per agent (annual) | | Professional | $60 per agent (annual) | | Executive | $200 per agent (annual) | | Super | $800 per agent (annual) | - **Usage:** Flat per agent; Day.ai says it has no usage pricing. Volume discounts. - **Free tier or trial:** Free tier that includes capture. ## Side by side | | Day.ai | Closefinch | |---|---|---| | Published prices | Yes, on day.ai/pricing | Yes, on this site, in pricing.md and in pricing.json | | Priced per | Agent | Seat (Free covers up to 3 seats) | | Free plan or trial | Free tier that includes capture. | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | Flat per agent; Day.ai says it has no usage pricing. Volume discounts. | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together We have no Day.ai connector. You can move data in with CSV or spreadsheet import. ## What differs - Day.ai prices per agent with no usage charges; we price per seat with included recording hours and AI credits, and meter beyond them. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/folk # Closefinch and Folk Folk is a CRM. Its figures below come from its own pricing page, [folk.app/pricing](https://www.folk.app/pricing). Checked 2026-11-05. Product names belong to their owners. ## Folk's published pricing | Plan | Price | |---|---| | Standard | $24 per member (annual), $30 (monthly) | | Premium | $48 per member (annual), $60 (monthly) | | Enterprise | From $80 per member | - **Usage:** Email-finder, AI-field and research credits. - **Free tier or trial:** 2-week trial, no card. ## Side by side | | Folk | Closefinch | |---|---|---| | Published prices | Yes, on folk.app/pricing | Yes, on this site, in pricing.md and in pricing.json | | Priced per | Member (billed per workspace) | Seat (Free covers up to 3 seats) | | Free plan or trial | 2-week trial, no card. | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | Email-finder, AI-field and research credits. | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together We have no Folk connector. You can move data in with CSV or spreadsheet import. ## What differs - Folk's paid plans start below ours. Our Free plan covers up to 3 seats with no time limit. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/lightfield # Closefinch and Lightfield Lightfield is an AI CRM. Its figures below come from its own pricing page, [lightfield.com/pricing](https://lightfield.com/pricing). Checked 2026-11-05. Product names belong to their owners. ## Lightfield's published pricing | Plan | Price | |---|---| | Starter | $89 per user (monthly, up to 3 seats) | | Pro | $999 per workspace (annual) | | Growth | $2,999 per workspace (annual) | | Enterprise | Custom | - **Usage:** Pooled workspace credits; extra capture seat $200. - **Free tier or trial:** 14-day trial; no free tier. ## Side by side | | Lightfield | Closefinch | |---|---|---| | Published prices | Yes, on lightfield.com/pricing | Yes, on this site, in pricing.md and in pricing.json | | Priced per | User on Starter; workspace on Pro and Growth | Seat (Free covers up to 3 seats) | | Free plan or trial | 14-day trial; no free tier. | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | Pooled workspace credits; extra capture seat $200. | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together We have no Lightfield connector. You can move data in with CSV or spreadsheet import. ## What differs - Lightfield's larger plans are priced per workspace; ours are per seat. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/compare/gong # Closefinch and Gong Gong is a revenue intelligence platform. Its figures below come from its own pricing page, [gong.io/pricing](https://www.gong.io/pricing). Checked 2026-11-05. Product names belong to their owners. ## Gong's published pricing | Plan | Price | |---|---| | All plans | Quote only | - **Usage:** Not published. - **Free tier or trial:** Not published. ## Side by side | | Gong | Closefinch | |---|---|---| | Published prices | No, quote only | Yes, on this site, in pricing.md and in pricing.json | | Priced per | Not published | Seat (Free covers up to 3 seats) | | Free plan or trial | Not published. | Free plan for up to 3 seats, with no time limit | | Usage beyond the plan | Not published. | $1.25 per extra recorder hour; 5,000 AI credits for $50 (paid plans, once billing launches; on Free, use pauses at the allowance) | ## Using them together Closefinch imports Gong calls (brief, key points, next steps and transcript) every 30 minutes, so the CRM and agents can use them. ## What differs - Gong doesn't publish prices; ours are public on this site, in pricing.md and in pricing.json. - Our call coaching scorecards and forecasting are in the Pro plan. ## Closefinch - Plans: Free (up to 3 seats), Team $35 and Pro $79 per seat a month billed annually ($44 and $99 monthly). - Captures email, calendar, meetings, Slack, calls and support tickets into accounts, contacts and deals. - Agents run on schedules or events, and every outgoing action passes a policy engine and, by default, a person's approval. - Meeting recorder, live transcript and post-meeting follow-ups, plus imports from platform transcripts and other notetakers. - A public REST API with OpenAPI 3.1, a hosted MCP server and a CLI. - Two-step sign-in, which owners need for owner settings, and a tamper-evident audit log. --- Source: https://www.closefinch.com/docs # Developer docs Closefinch is built to be used by programs and AI agents as well as people. Everything below works with an account on any plan, including Free. ## REST API - **Spec:** OpenAPI 3.1 at [https://www.closefinch.com/openapi.json](https://www.closefinch.com/openapi.json). It lists every route, the scope each one needs, request and response schemas, and the rate limits. - **Base URL:** `https://api.closefinch.com/v1` - **Authentication:** `Authorization: Bearer ra_live_…`. An owner creates keys in Settings → API keys and chooses their scopes. A key never has more access than the person who made it, and it stops working when they leave. Details: [Authentication](https://www.closefinch.com/docs/authentication). - **IDs** are opaque and prefixed (`acct_…`, `ct_…`, `deal_…`). Lists use cursor pagination. - **Errors** say what went wrong and what to do next: `{ "error": { "code": "…", "message": "…" } }`. - **Writes that send something** (email, CRM updates in other systems) go through the same policy engine and approval inbox as the app. ## MCP server Connect Claude, ChatGPT, Cursor or any MCP client to `https://api.closefinch.com/mcp` (Streamable HTTP). Clients sign in with OAuth, registering themselves, or send an API key. Details: [MCP server](https://www.closefinch.com/docs/mcp). ## CLI The `ra` command line covers every resource, signs in with OAuth profiles or a key, prints JSON, and has a built-in `--jq`. `ra install claude-code` writes a project's `.mcp.json` and a skill for Claude Code; it references `${RA_API_KEY}`, never the key itself. Install and every command: [CLI](https://www.closefinch.com/docs/cli). ## Webhooks - **Outbound:** subscribe to record and agent events. Every delivery is signed per Standard Webhooks (`webhook-id`, `webhook-timestamp`, `webhook-signature`). - **Inbound:** start an agent from any system with a webhook trigger, authenticated by the trigger's secret (`Authorization: Bearer …`). Zapier, Make and n8n work too. Signing, retries and the event list: [Webhooks](https://www.closefinch.com/docs/webhooks). ## Machine-readable files - [llms.txt](https://www.closefinch.com/llms.txt) and [llms-full.txt](https://www.closefinch.com/llms-full.txt) - [pricing.json](https://www.closefinch.com/pricing.json) and [pricing.md](https://www.closefinch.com/pricing.md) - Every page has a Markdown version: add `.md` to its path (`/pricing.md`, `/index.md` for the home page) or send `Accept: text/markdown`. - [security.txt](https://www.closefinch.com/.well-known/security.txt) --- Source: https://www.closefinch.com/docs/mcp # MCP server Closefinch runs a hosted [Model Context Protocol](https://modelcontextprotocol.io) server, so assistants can read and work in your CRM with your permissions. - **Endpoint:** `https://api.closefinch.com/mcp` (Streamable HTTP, JSON responses; protocol versions 2025-06-18 and 2025-03-26). - **OAuth sign-in.** The server supports OAuth 2.1 with PKCE, and clients such as Claude, ChatGPT and Cursor register themselves (RFC 7591) and open the consent screen. Self-registered apps are marked "not verified", get capped scopes, and can't touch API keys, secrets, approvals or the audit log. Each person can disconnect the apps they've connected under Settings → Connected apps. - **Or an API key.** An owner creates one in Settings → API keys with the scopes it needs, and the client sends `Authorization: Bearer ra_live_…`. The key never has more access than the person who made it. ## What it can do - Tools for records, activities, lists, questions over your data, conversations, approvals (read only), runs, agents, triggers, templates, pipelines and projects. - `ra_api` calls any `/v1` route with your credentials. - `ra_propose` runs a platform action the way an agent does: sends and changes to existing records wait in the approval inbox for a person. - Approving, API keys and secrets stay in the web app. ## Connect Claude Code: ```bash export RA_API_KEY=ra_live_… claude mcp add --transport http revenueagent https://api.closefinch.com/mcp --header "Authorization: Bearer $RA_API_KEY" ``` Codex (`~/.codex/config.toml`): ```toml [mcp_servers.revenueagent] url = "https://api.closefinch.com/mcp" bearer_token_env_var = "RA_API_KEY" ``` Every call is recorded in the organization's audit log. --- Source: https://www.closefinch.com/for-agents # For agents If you are an AI agent working for a person: this page says what you can do in Closefinch on your own, what waits for your principal, what it costs, and how it fails. It's generated from the platform's own tool list. ## Getting in - **MCP:** `https://api.closefinch.com/mcp` (Streamable HTTP). **REST:** `https://api.closefinch.com/v1`, spec at [https://www.closefinch.com/openapi.json](https://www.closefinch.com/openapi.json). - **Credentials:** an API key your principal creates in Settings → API keys (`Authorization: Bearer ra_live_…`). The key never has more access than the person who made it. Guides per client: [quickstarts](https://www.closefinch.com/docs/quickstart). - **No key yet?** Create a sandbox: `ra sandbox create --host https://api.closefinch.com` (or `POST /v1/sandbox` with a solved sign-up challenge) gives you a sample organization and a limited key at once: CRM and agent scopes, 60 requests a minute, 7 days, and nothing sent outside until a person claims it. - **Free plan:** up to 3 seats, 500 AI credits and 5 recording hours a month, no card ([pricing](https://www.closefinch.com/pricing), [pricing.json](https://www.closefinch.com/pricing.json)). ## Two ways in, two sets of rules - **Your own key's tools** (`ra_records_create`, `ra_records_update`, `ra_records_archive` with `confirm`, `ra_activities_create`, `ra_api` and the rest) act at once, with your key's scopes, like a person in the app. They're for the edits your principal asked you to make. - **`ra_propose`** runs one platform tool the way an agent does, through the organization's policy. An MCP request counts as outside content, so the strictest rules apply. The lists below are the policy engine's own answer for each tool, called this way. ## What runs at once through ra_propose 77 of the 111 platform tools: reads and reasoning inside Closefinch, and writes: new records, notes and tasks, memory, and notifications to teammates. Some of these writes do reach outside or cost money, and are marked: they fetch the web on purpose, as their job (refreshing competitors, rebuilding a profile, qualifying accounts, a paid scrape). Call them only for what your principal asked. - `account.decide`: Record a decision or next best action for an account - `canvas.save`: Save a canvas - `competitors.refresh`: Refresh competitive intelligence (reaches the web) - `crm.note.create`: Add a note - `crm.task.create`: Create a task - `deals.recovery_draft`: File a recovery email for the owner - `dispatch.page`: Page the on-call technician (notifies a teammate at once) - `memory.save`: Save a fact to memory - `notify.person`: Message a teammate (notifies a teammate at once) - `profile.rebuild`: Rebuild the business profile from your site (reaches the web) - `proof.mine`: Mine closed-won deals for proof points - `qualify.accounts`: Qualify accounts against fit questions (reaches the web) - `recruit.next_batch`: Propose the next design-partner outreach - `segment.stage`: Stage a segment - `segments.build_from_icp`: Build segments from the ICP - `web.scrape_apify`: Scrape with an Apify actor (paid) (reaches the web) (paid per call) - and these reads: `account.plan`, `activities.search`, `ads.google.query`, `ads.linkedin.report`, `ads.reddit.report`, `analytics.ga4.report`, `background.shared`, `committee.get`, `cpg.account_health`, `cpg.action_list`, `cpg.category_review`, `crm.analyze_write`, `crm.record.get`, `crm.records.search`, `customers.handoff_note`, `deals.review`, `deals.stalled`, `dispatch.triage`, `equipment.account_open`, `equipment.fleet_review`, `forecast.read`, `fundraising.pipeline_review`, `glossary.lookup`, `graph.neighborhood`, `home_services.estimates_review`, `home_services.oncall_review`, `home_services.triage`, `icp.check`, `industrial.rfqs`, `industrial.slow_movers`, `kb.read`, `knowledge.search`, `meetings.brief`, `meetings.followthrough`, `meetings.status`, `memory.list`, `personas.match`, `phones.best`, `phones.list`, `pipeline.stalled_deals`, `prospects.match`, `quotes.get`, `quotes.intel`, `quotes.pipeline`, `relationships.lookup`, `sandbox.run`, `search_console.query`, `signals.list`, `signals.why_now`, `sms.draft_reply`, `tam.overview`, `transcripts.read`, `voice.call_review`, `voice.caller_context`, `voice.campaign_progress`, `voice.catalog`, `voice.find_times`, `voice.payment_status`, `voice.score`, `voice.verify_caller`, `warm.paths` ## What waits for a person These wait in the approval inbox, and `ra_propose` returns the approval's link for your principal: anything that reaches outside (web pages, other MCP servers, the browser), sends, and changes to existing records. - `account.plan.refresh`: Refresh the account plan (changes an existing record) - `ads.campaign_status`: Pause or enable an ad campaign (sends outside Closefinch) (changes an existing record) - `agentmail.send`: Send from the agent's mailbox (sends outside Closefinch) - `browser.act`: Click or type in your browser (sends outside Closefinch) (changes an existing record) - `browser.extract_table`: Read a table in your browser (reads outside Closefinch) - `browser.read`: Read a page in your browser (reads outside Closefinch) - `committee.build`: Map the buying committee (changes an existing record) - `company.changes`: Hiring, leadership and tech changes for an account (reads outside Closefinch) - `crm.record.update`: Update a record (changes an existing record) - `http.get`: Fetch a URL (reads outside Closefinch) - `http.send`: Send to a URL (sends outside Closefinch) - `mcp.call`: Act through a connected MCP server (sends outside Closefinch) - `mcp.read`: Read from a connected MCP server (reads outside Closefinch) - `media.screenshot`: Screenshot a public web page (sends outside Closefinch) - `outbound.dnc.add`: Add to the do-not-contact list (changes an existing record) - `records.account`: Public records for an account (reads outside Closefinch) - `research.brief`: Research an account or person (reads outside Closefinch) - `sandbox.run_net`: Run code (allowlisted network) (reads outside Closefinch) - `segment.append`: Add to a segment (changes an existing record) - `sms.handoff`: Hand a text thread to a person (changes an existing record) - `voice.book_meeting`: Book a meeting (sends outside Closefinch) - `web.read`: Read a web page (reads outside Closefinch) - `web.scrape`: Scrape structured data from web pages (reads outside Closefinch) - `web.search`: Search the web or news (reads outside Closefinch) ## Checked per call These are decided for each call, against facts only the call has (who it reaches, the number's consent and quiet hours). Without them they're refused: - `sms.reply`: Answer a text ## For live calls and texts only These belong to voice agents and the texting desk: they act on the person on a live call or thread, need that call, and aren't for an MCP agent: `voice.campaign_enroll`, `voice.campaign_result`, `voice.payment_link`, `voice.save_caller`, `voice.send_booking_confirmation`, `voice.send_sms`, `voice.speed_to_lead`, `voice.take_message`, `voice.warm_transfer`. Approving is never possible through the API or MCP: it happens in the web app or Slack. See [approvals and policy](https://www.closefinch.com/docs/approvals). ## What it costs | Unit | Closefinch | Notes | |---|---|---| | An API or MCP call | No charge per call | Rate-limited (below) | | AI work (summaries, drafts, agent runs) | $0.01 per credit (5,000 for $50) | Each run's trace shows its cost; a dry run (`ra_runs_start` with `dry_run: true`) shows the plan first | | A bot-recorded meeting hour beyond the plan | $1.25 | Imported Zoom, Meet or Teams transcripts use no hours | | A seat | Team $35, Pro $79 a month, billed annually | Agents aren't seats | What others publish for the same kind of usage (checked on the date shown, from their own pages): | Product | Published usage pricing | Checked | |---|---|---| | Attio | Credits per seat per month: 100 / 500 / 1,000 / 2,500 by plan. Packs from 5,000 for $70 to 50,000 for $475 (annual). | 2026-11-05 | | HubSpot Sales Hub | Credits per month: 500 / 3,000 / 5,000 by plan; extra credits $0.01 each. | 2026-11-05 | | Close | AI credits 500 to 2,000 a month by plan; calls about $0.02 a minute. | 2026-11-05 | | Day.ai | Flat per agent; Day.ai says it has no usage pricing. Volume discounts. | 2026-11-05 | | Folk | Email-finder, AI-field and research credits. | 2026-11-05 | | Lightfield | Pooled workspace credits; extra capture seat $200. | 2026-11-05 | ## Paying You can't pay, and you don't need to. To suggest a paid plan, call `ra_billing_upgrade_link` (or `POST /v1/billing/upgrade-link`). It returns a link to the billing page with the plan filled in and tells the organization's owners. An owner reviews it and pays through checkout, and nothing is bought until then. Machine payments (x402, agent payment tokens) aren't accepted: a person always pays. ## Limits and errors - **Rate limit:** per key, published in `X-RateLimit-Limit`, `X-RateLimit-Remaining` and `X-RateLimit-Reset`. Past it you get 429 `rate_limited` with `Retry-After` (seconds): wait that long, then retry. - **Errors** are `{ "error": { "code", "message", "fix", "docs" } }`: the cause, what to do, and a link to its section of [the errors page](https://www.closefinch.com/docs/errors). - `bad_request` (400): the input didn't validate; `details.issues` lists each field and what's wrong. Fix the input and send it again. - `unauthorized` (401): no key or a bad one. Ask your principal for a key from Settings → API keys. - `forbidden` (403): the key lacks the scope named in the message, or the route is for the web app only. - `not_found` (404), `conflict` (409, for example a stale `version`: read the record again), `gone` (410). - `policy_blocked` (422): the organization's policy refused it (the kill switch, consent, do-not-contact). Don't retry: tell your principal which rule. - `unavailable` (503): retry after a short wait. - Archiving needs `confirm: true`. A merged record answers `moved` (301) with the record it was merged into in `Location`: read that one instead. ## A snippet for AGENTS.md, CLAUDE.md or Cursor rules ```markdown ## Closefinch (CRM) - MCP server: https://api.closefinch.com/mcp with `Authorization: Bearer $RA_API_KEY` (never write the key into a file). - Start with ra_whoami to see the organization and scopes. - Read freely. Your key's own tools (ra_records_update, ra_api…) act at once: use them only for what the user asked. - For sends, outside requests and changes the user didn't ask for, use ra_propose: they wait for a person's approval. Never try to approve. - On 429, wait Retry-After seconds. On policy_blocked, stop and report the rule to the user. - Docs: https://www.closefinch.com/docs (every page also as .md); tools: https://www.closefinch.com/docs/mcp/tools. ``` --- Source: https://www.closefinch.com/docs/quickstart # Quickstarts Pick your client. Each guide lists what you need, the one command or file to add, and how to check it works. Everything runs on any plan, including Free. - [Claude Code](https://www.closefinch.com/docs/quickstart/claude-code) - [Codex](https://www.closefinch.com/docs/quickstart/codex) - [Cursor](https://www.closefinch.com/docs/quickstart/cursor) - [ChatGPT](https://www.closefinch.com/docs/quickstart/chatgpt) - [Claude Agent SDK](https://www.closefinch.com/docs/quickstart/agent-sdk) - [REST API](https://www.closefinch.com/docs/quickstart/rest) What you need for all of them: an account and an API key (Settings → API keys). The MCP server is at `https://api.closefinch.com/mcp` (Streamable HTTP); the REST API is at `https://api.closefinch.com/v1` with its spec at [https://www.closefinch.com/openapi.json](https://www.closefinch.com/openapi.json). --- Source: https://www.closefinch.com/docs/quickstart/claude-code # Quickstart: Claude Code **You need:** Claude Code, and an API key in `RA_API_KEY`. 1. **Get a key.** Sign up at https://app.closefinch.com/sign-up (the Free plan, no card). An owner creates a key in Settings → API keys and picks its scopes: `crm:read` to read, plus `crm:write` and `agents:write` to make changes. Keep it in an environment variable, never in a file you commit: ```bash export RA_API_KEY=ra_live_… ``` 2. **Add the server** (for this project; use `--scope user` for all your projects): ```bash claude mcp add --transport http revenueagent https://api.closefinch.com/mcp --header "Authorization: Bearer $RA_API_KEY" ``` The shell fills in `$RA_API_KEY`, so Claude Code stores the key itself in `~/.claude.json`: keep it at local or user scope, never `--scope project` (that file is shared). To keep only a reference, use the `ra` CLI instead: `ra install claude-code` writes the project's `.mcp.json` and a skill, which refer to `${RA_API_KEY}` and never contain the key. 3. **Check it.** Check it works: ask the assistant "Who am I in Closefinch?" It calls `ra_whoami` and answers with your organization, role and scopes. Next: [task recipes](https://www.closefinch.com/docs/recipes), [the MCP tools](https://www.closefinch.com/docs/mcp/tools), and [what an agent can do alone](https://www.closefinch.com/for-agents). --- Source: https://www.closefinch.com/docs/quickstart/codex # Quickstart: Codex **You need:** the Codex CLI or IDE extension, and an API key in `RA_API_KEY`. 1. **Get a key.** Sign up at https://app.closefinch.com/sign-up (the Free plan, no card). An owner creates a key in Settings → API keys and picks its scopes: `crm:read` to read, plus `crm:write` and `agents:write` to make changes. Keep it in an environment variable, never in a file you commit: ```bash export RA_API_KEY=ra_live_… ``` 2. **Add the server** to `~/.codex/config.toml`: ```toml [mcp_servers.revenueagent] url = "https://api.closefinch.com/mcp" bearer_token_env_var = "RA_API_KEY" ``` 3. **Check it.** Check it works: ask the assistant "Who am I in Closefinch?" It calls `ra_whoami` and answers with your organization, role and scopes. Next: [task recipes](https://www.closefinch.com/docs/recipes) and [the MCP tools](https://www.closefinch.com/docs/mcp/tools). --- Source: https://www.closefinch.com/docs/quickstart/cursor # Quickstart: Cursor **You need:** Cursor, and an API key in `RA_API_KEY`. 1. **Get a key.** Sign up at https://app.closefinch.com/sign-up (the Free plan, no card). An owner creates a key in Settings → API keys and picks its scopes: `crm:read` to read, plus `crm:write` and `agents:write` to make changes. Keep it in an environment variable, never in a file you commit: ```bash export RA_API_KEY=ra_live_… ``` 2. **Add the server** to `.cursor/mcp.json` in the project (or `~/.cursor/mcp.json` for every project): ```json { "mcpServers": { "revenueagent": { "url": "https://api.closefinch.com/mcp", "headers": { "Authorization": "Bearer ${env:RA_API_KEY}" } } } } ``` 3. **Check it.** Open Settings → MCP and confirm the server shows its tools, then: Check it works: ask the assistant "Who am I in Closefinch?" It calls `ra_whoami` and answers with your organization, role and scopes. Next: [task recipes](https://www.closefinch.com/docs/recipes) and [a snippet for Cursor rules](https://www.closefinch.com/for-agents#a-snippet-for-agentsmd-claudemd-or-cursor-rules). --- Source: https://www.closefinch.com/docs/quickstart/chatgpt # Quickstart: ChatGPT **You need:** a ChatGPT plan that allows custom connectors (developer mode), and a Closefinch account. 1. **Add the connector.** In ChatGPT, open Settings → Apps and connectors → Advanced, turn on developer mode, and create a connector with the server URL `https://api.closefinch.com/mcp`. 2. **Sign in.** ChatGPT registers itself (RFC 7591) and opens Closefinch's consent screen. You choose what it may do. Self-registered apps are marked "not verified" and get capped scopes; you can disconnect it under Settings → Connected apps. ChatGPT can't connect with a key; for a key, use [Claude Code](https://www.closefinch.com/docs/quickstart/claude-code), [Codex](https://www.closefinch.com/docs/quickstart/codex) or [Cursor](https://www.closefinch.com/docs/quickstart/cursor). 3. **Check it.** In a chat with the connector on, ask "Who am I in Closefinch?" Approving actions stays in the Closefinch app or Slack, never in the chat. --- Source: https://www.closefinch.com/docs/quickstart/agent-sdk # Quickstart: Claude Agent SDK **You need:** Node.js 20 or later, `@anthropic-ai/claude-agent-sdk` (pin the version you install), an Anthropic API key in `ANTHROPIC_API_KEY`, and a Closefinch key in `RA_API_KEY`. 1. **Get a key.** Sign up at https://app.closefinch.com/sign-up (the Free plan, no card). An owner creates a key in Settings → API keys and picks its scopes: `crm:read` to read, plus `crm:write` and `agents:write` to make changes. Keep it in an environment variable, never in a file you commit: ```bash export RA_API_KEY=ra_live_… ``` 2. **Install** (pinned): ```bash npm install --save-exact @anthropic-ai/claude-agent-sdk ``` 3. **Give the agent the server:** ```ts import { query } from "@anthropic-ai/claude-agent-sdk"; for await (const message of query({ prompt: "Which of my deals close this month and have no next step?", options: { mcpServers: { revenueagent: { type: "http", url: "https://api.closefinch.com/mcp", headers: { Authorization: `Bearer ${process.env.RA_API_KEY}` } }, }, allowedTools: ["mcp__revenueagent__ra_records_list", "mcp__revenueagent__ra_ask"], }, })) { if (message.type === "result") console.log(message.result); } ``` 4. **Check it.** The answer lists deals from your account. Any send or change to an existing record goes through `ra_propose` and waits in the approval inbox. Next: [task recipes](https://www.closefinch.com/docs/recipes), [the MCP tools](https://www.closefinch.com/docs/mcp/tools), and [what an agent can do alone](https://www.closefinch.com/for-agents). --- Source: https://www.closefinch.com/docs/quickstart/rest # Quickstart: REST API **You need:** an API key in `RA_API_KEY` and any HTTP client. 1. **Get a key.** Sign up at https://app.closefinch.com/sign-up (the Free plan, no card). An owner creates a key in Settings → API keys and picks its scopes: `crm:read` to read, plus `crm:write` and `agents:write` to make changes. Keep it in an environment variable, never in a file you commit: ```bash export RA_API_KEY=ra_live_… ``` 2. **Call it** (no key yet? `ra sandbox create --host https://api.closefinch.com` makes a sandbox and saves its key): ```bash curl -s https://api.closefinch.com/v1/me -H "Authorization: Bearer $RA_API_KEY" curl -s "https://api.closefinch.com/v1/deals?q=Acme&limit=10" -H "Authorization: Bearer $RA_API_KEY" curl -s https://api.closefinch.com/v1/tasks -H "Authorization: Bearer $RA_API_KEY" -H "content-type: application/json" \ -d '{"values":{"title":"Send the pricing sheet"}}' ``` 3. **Check it.** `/v1/me` answers with your organization and scopes. The spec for every route is at [https://www.closefinch.com/openapi.json](https://www.closefinch.com/openapi.json), and [the reference](https://www.closefinch.com/docs/api) lists them by resource. Errors are `{ "error": { "code", "message", "details" } }`; see [limits and errors](https://www.closefinch.com/for-agents#limits-and-errors). --- Source: https://www.closefinch.com/docs/recipes # Task recipes Each recipe names the tools in the order an agent calls them. The same calls work over REST ([reference](https://www.closefinch.com/docs/api)). ## Prepare for a meeting 1. `ra_records_list` with `object: "accounts"` and `query.q` set to the company, to get its id. 2. `ra_account_state` for its memory, open decisions and the next best action. 3. `ra_records_timeline` for the latest emails, meetings and notes, each with its source. 4. `ra_graph_neighborhood` for who's involved and how recently they talked. ## Find stalled deals and add follow-ups 1. `ra_ask`: "Which open deals have had no activity in 14 days?" (it answers with the records it used). 2. For each one, `ra_propose` with `crm.task.create` and the deal in `related`. Creating a task happens at once and can be undone in the app. ## Log a call `ra_activities_create` with `type: "call"`, the time, a summary and the contact's id in `record_ids`. It joins the contact's and the account's timelines. ## Change a record, with approval `ra_propose` with `crm.record.update`: the change waits in the approval inbox, and the result has the approval's link for a person to decide. Direct `ra_records_update` is for your own edits, with your key's scopes. ## Check an agent before it's live 1. `ra_agents_draft_from_goal` with the goal in plain language: a draft, not a live agent. 2. `ra_agents_test_run`: a dry run (nothing is sent or changed), with the trace. 3. `ra_evals_add` and `ra_evals_run` to keep it honest after edits. 4. Publish it with `ra_agents_publish`. Any key with `agents:write` can publish, so keep that scope off keys an agent holds unless your principal wants it to put agents live. ## Answer a security questionnaire `ra_questionnaire_answer` with the questions: answers come from approved sources only, each with its citation. --- Source: https://www.closefinch.com/docs/authentication # Authentication Every request to `https://api.closefinch.com/v1` and the MCP server at `https://api.closefinch.com/mcp` carries a bearer token: an API key, or an OAuth access token for an app or the CLI. ```bash curl -s https://api.closefinch.com/v1/me -H "Authorization: Bearer $RA_API_KEY" ``` ## API keys - **Who makes them:** an owner, in Settings → API keys, choosing the scopes the key needs (`crm:read`, `crm:write`, `agents:write` and so on). - **What they can do:** never more than the person who made it, and a key stops working when that person leaves. - **How long they last:** 90 days by default, up to 365. A key that never expires is an owner's explicit choice, with a reason, and it's recorded. The creator is reminded 14 and 3 days before a key expires. - **Where they work from:** optionally, only from the addresses you list (an IP allowlist per key), and with a rate limit of its own. - **How they're kept:** shown once when created, and stored only as a hash. ## OAuth for apps and MCP clients Apps and assistants sign in with OAuth 2.1 and PKCE. MCP clients such as Claude, ChatGPT and Cursor register themselves (RFC 7591) and open the consent screen; self-registered apps are marked "not verified", get capped scopes, and can't touch API keys, secrets, approvals or the audit log. Each person can disconnect the apps they've connected under Settings → Connected apps. The metadata is at `https://api.closefinch.com/.well-known/oauth-protected-resource/mcp`. Details: [MCP server](https://www.closefinch.com/docs/mcp). ## The CLI `ra login --host https://api.closefinch.com` signs in with the OAuth device flow (RFC 8628): it shows a short code, you approve it in the browser, and there's no key to paste. Scripts and CI can use a key instead (`RA_HOST`, `RA_API_KEY`). Details: [CLI](https://www.closefinch.com/docs/cli). ## What a token can't do Deciding approvals happens in the web app (https://app.closefinch.com) or in Slack, never with an API key or an app token. API keys are created, changed and rolled only in the web app, and so are workflow secrets. A credential with the `keys:write` scope can list and revoke keys (useful in an incident), so give that scope only to what needs it. Errors for missing scopes, expired keys and blocked addresses are in [Errors](https://www.closefinch.com/docs/errors). --- Source: https://www.closefinch.com/docs/cli # CLI `ra` is Closefinch's command line, built for people and coding agents. It covers every resource in the REST API, prints a table in a terminal and JSON when piped, and has a built-in `--jq`. It has no dependencies (Node's standard library only) and needs Node 22.12 or newer. Version 0.1.0. ## Install It isn't on the npm registry yet: install the package built from the repository, with npm. Never pipe a script into a shell. ```bash node tools/cli/pack.mjs # in the repository: writes apps/cli/dist/revenueagent-cli-0.1.0.tgz npm install -g ./apps/cli/dist/revenueagent-cli-0.1.0.tgz ra --version ``` ## Sign in ```bash ra login --host https://api.closefinch.com ``` It shows a short code; approve it in your browser (OAuth device flow, nothing to paste). For scripts, use a key from Settings → API keys instead, read from a variable rather than typed on the command line: `printf %s "$RA_API_KEY" | ra auth login --host https://api.closefinch.com --with-key`. In CI, set `RA_HOST` and `RA_API_KEY`. `ra auth status` shows who you are; `ra auth logout` forgets the profile. ## Common commands ```bash ra deals list --query q=Acme --fields id,name ra deals get deal_… ra tasks create --field values.title="Call back" --raw-field 'values.related=["deal_…"]' ra api GET /v1/forecast ra runs start agt_… --input '{"deal_id":"deal_…"}' --wait ra bulk-create contacts --input contacts.jsonl ``` - Deleting and merging need `--force`; `--dry-run` prints the request instead of sending it. - `ra runs start … --wait` exits 1 if the run failed; 3 if it's waiting for a person, paused or sleeping; and 0 otherwise: done, cancelled, or still running after about 20 minutes of waiting (check its status before relying on it). - `ra agent-context` prints, as JSON, what the CLI and your credential can do. ## Filter with --jq ```bash ra deals list --jq '.data[] | select(.values.amount.amount > 10000) | {id, name}' ``` A deal's amount is `{ amount, currency }`, so compare `.values.amount.amount`. The built-in jq covers paths, `select`, `map`, comparisons, object and array construction, string interpolation, `if`/`reduce`/`try`, `sort_by`/`group_by`/`unique_by`, `to_entries`/`with_entries`, regular expressions and formats such as `@csv` and `@json`. It only reads: no assignment, no `def`, no environment access. ## Set up Claude Code ```bash ra install claude-code ``` It adds the MCP server to the project's `.mcp.json` and a skill for Claude Code. They reference `${RA_API_KEY}`, never the key itself, so set `RA_API_KEY` to a key from Settings → API keys (the browser sign-in doesn't make one). `ra install codex` prints the block for Codex's config instead. Quickstarts for other clients: [Quickstarts](https://www.closefinch.com/docs/quickstart). ## Every command From `ra help` (version 0.1.0): ```text ra 0.1.0: the Closefinch CLI ra login --host [--scope "crm:read …"] sign in in your browser with a short code (no key to paste) ra auth login --host --key sign in with an API key (or --with-key to read it from stdin) ra auth login --host --client-id sign in with OAuth (an owner registers the app with redirect http://127.0.0.1:8765/callback; --port to change) ra auth status | ra auth logout ra sandbox create --host [--agent-name ] [--claim-email ] for agents: a sandbox organization with sample data and a limited key (7 days, no outbound until a person claims it), saved as a profile ra api [--field k=v] [--raw-field k=] [--body |@file|-] [--query k=v] ra list [--query k=v] [--limit n] [--all] ra get | create [--field …] | update [--field …] | delete --force ra bulk-create --input [--batch ] 25 per request; exits 1 if any row fails ra runs start [--input ] [--wait] ra agent-context what this CLI and your credential can do, as JSON ra install claude-code [--dir .] add the MCP server (.mcp.json) and a skill for Claude Code ra install codex print the Codex config block (writes nothing) Output: a table in a terminal, JSON when piped. --json, --output table|json|toon, --fields a,b.c, --jq (a built-in jq). --dry-run prints the request instead of sending it. --profile , --host . CI: RA_HOST, RA_API_KEY. ``` --- Source: https://www.closefinch.com/docs/webhooks # Webhooks ## Outbound: events sent to you Add an endpoint in Settings → Webhooks (or `POST /v1/webhooks`) and choose the events it receives, or `*` for all. The signing secret (`whsec_…`) is shown once. Endpoints must use https and a public address. Events: `account.dormant`, `account.declining`, `nps.answered`, `invoice.overdue`, `record.created`, `record.updated`, `record.archived`, `deal.won`, `deal.lost`, `deal.stage_changed`, `activity.created`, `meeting.recorded`, `meeting.upcoming`, `insight.captured`, `run.finished`, `approval.needed`, `account.state_changed`, `signal.fired`, `sequence.replied`, `booking.no_show`, `quote.stalled`, `call.completed`. Each delivery is a POST with a JSON body, signed per [Standard Webhooks](https://www.standardwebhooks.com): - `webhook-id`: the event's id (use it to drop duplicates); - `webhook-timestamp`: seconds since 1970; - `webhook-signature`: `v1,` and the base64 HMAC-SHA256 of `id.timestamp.body`, keyed with the secret after `whsec_`. During the 24 hours after you rotate the secret, both secrets sign, separated by a space. ```js import { createHmac, timingSafeEqual } from "node:crypto"; // body: the raw request body, as a string, before any JSON parsing. function verify(secret, headers, body) { const id = headers["webhook-id"], ts = headers["webhook-timestamp"], sigs = headers["webhook-signature"]; if (!id || !ts || !sigs) return false; const t = Number(ts); if (!Number.isFinite(t) || Math.abs(Date.now() / 1000 - t) > 300) return false; // more than five minutes off const key = Buffer.from(secret.replace(/^whsec_/, ""), "base64"); const want = createHmac("sha256", key).update(`${id}.${ts}.${body}`).digest(); return sigs.split(" ").some((sig) => { const got = Buffer.from(sig.replace(/^v1,/, ""), "base64"); return got.length === want.length && timingSafeEqual(got, want); }); } ``` Answer with any 2xx within 15 seconds. A failed delivery is tried again after 5 seconds, 5 minutes, 30 minutes, 2 hours, 5 hours, 10 hours, 10 hours (7 more attempts), then marked failed. Deliveries are kept for 30 days (`GET /v1/webhooks/{id}/deliveries`) and can be sent again (`POST /v1/webhook-deliveries/{id}/resend`). Pause an endpoint with `enabled: false`. ## Inbound: start an agent from another system Give an agent a webhook trigger (`POST /v1/agents/{id}/triggers`, or in the agent's settings). Its URL and secret are shown once. The other system posts JSON to that URL with `Authorization: Bearer `. The trigger maps the body's fields to the agent's inputs; the body is data for the run, never instructions. Send an `Idempotency-Key` header so a retried call doesn't start a second run. Zapier, Make and n8n work the same way. Recent firings, including duplicates and ones refused because the agent isn't live, are at `GET /v1/triggers/{id}/fires`; `POST /v1/triggers/{id}/rotate-secret` replaces the secret. Approvals, limits and what an agent may do alone: [Approvals and policy](https://www.closefinch.com/docs/approvals). --- Source: https://www.closefinch.com/docs/approvals # How approvals and policy work Every write that leaves Closefinch (an email, a text, a call, an update in another CRM, a Slack post) goes through the organization's policy before it runs. Agents, the API, MCP clients and people in the app meet the same rules. ## What happens at once, and what waits - **At once:** reading, reasoning, and creating new records inside Closefinch (tasks, notes, contacts). These can be undone in the app. - **Waits for a person:** changes to existing records, anything sent outside, and anything destructive, at the default autonomy. A run that read outside content (an MCP request counts) also waits before it fetches anything outside. The item lands in the approval inbox with what would happen, why, and its evidence. - **A key's own calls** (the REST API, and MCP's record and `ra_api` tools) act at once with the key's scopes, as a person in the app would. Only agent runs and `ra_propose` go through approvals. - **Deciding** happens in the web app or Slack, never through the API or MCP. An agent can list and read what's waiting (`ra_approvals_list`, `ra_approvals_get`) but not approve it. - An owner can give an agent more autonomy, one action type at a time, under Settings → Agent governance. ## The policy - **The kill switch:** an owner can stop all outbound at once (Settings → AI and outbound). Sends are then refused with `policy_blocked` (HTTP 422). - **Consent and do-not-contact:** opt-outs, the do-not-contact list, quiet hours and channel rules are checked before a send. A refusal says which rule. - **Budgets:** daily and monthly AI budgets pause runs when reached, and every run's trace shows its cost. - **Audit:** every call, by a person, a key or an MCP client, is in the organization's audit log. The full list of what an agent can do alone is on [the page for agents](https://www.closefinch.com/for-agents). --- Source: https://www.closefinch.com/docs/errors # Errors Every error from the REST API and the MCP server says what went wrong, what to do, and links here: ```json { "error": { "code": "unauthorized", "message": "…", "fix": "…", "docs": "https://www.closefinch.com/docs/errors#unauthorized" } } ``` Over MCP, a JSON-RPC error carries the same `fix` and `docs` in `error.data`, and a tool's error result in its content. Rate limits are in `X-RateLimit-Limit`, `X-RateLimit-Remaining` and `X-RateLimit-Reset` on every response. ## bad_request **HTTP 400: The request didn't validate.** Fix what the message (and details.issues, when present) names, then send it again. ## unauthorized **HTTP 401: No credential, or a bad one.** Send Authorization: Bearer . Get a key in Settings → API keys, or create a sandbox with its own key: POST /v1/sandbox (or `ra sandbox create`). ## forbidden **HTTP 403: Not allowed with this credential.** Use a key with the scope the message names (Settings → API keys), or do this in the web app if it's web-only. ## not_found **HTTP 404: No such thing.** Check the id or the path. List the resource to find the id, or see /v1/openapi.json for the routes. ## conflict **HTTP 409: It changed, or it can't happen in this state.** Read it again and retry with its current version or state. ## moved **HTTP 301: Merged into another record.** Read the record at the Location header (details.merged_into) instead. ## gone **HTTP 410: No longer available.** It was removed or expired; don't retry. ## rate_limited **HTTP 429: Too many requests.** Wait the number of seconds in Retry-After, then retry. X-RateLimit-Remaining shows what's left. ## policy_blocked **HTTP 422: The organization's policy refused it.** Don't retry: tell a person which rule the message names (kill switch, consent, do-not-contact, budgets). ## unavailable **HTTP 503: Temporarily unavailable.** Retry after a short wait (the Retry-After header, when present). ## internal **HTTP 500: Something failed on our side.** Retry once. If it fails again, report the request id (the x-request-id header). No key yet? Create a sandbox with its own limited key: [quickstart](https://www.closefinch.com/docs/quickstart/rest) or `ra sandbox create`. --- Source: https://www.closefinch.com/docs/changelog # Changelog The API and the MCP server share one version. A breaking change gets a new version, and a deprecation notice here with its date before it takes effect. The spec for the current version is [https://www.closefinch.com/openapi.json](https://www.closefinch.com/openapi.json). What's shipped in the product itself is in the [product changelog](https://www.closefinch.com/changelog). ## 0.1.0 (2026-11-06) - Every API and MCP error carries `fix` and `docs` (a link to its section of /docs/errors); a missing key says where to get one. - `ra sandbox create`: an agent gets a sandbox organization and a limited key from the CLI, saved as a profile. - `ra_billing_upgrade_link` / `POST /v1/billing/upgrade-link`: an agent suggests a paid plan; an owner reviews and pays. - MCP tool descriptions are written task first, each with an example call. - `ra_templates_list` lists agent templates again (its route is `/v1/agent-templates`). - Developer docs: quickstarts per client, task recipes, approvals and policy, and a reference page per MCP tool and API resource. - MCP server: protocol versions 2025-06-18 and 2025-03-26. **Deprecations:** none. --- Source: https://www.closefinch.com/docs/mcp/tools # MCP tools Every tool the server at `https://api.closefinch.com/mcp` offers, with what it's for. Each has its own page with an example call. Read-only tools change nothing; the others say what they change. | Tool | What it's for | Changes data | |---|---|---| | [`ra_account_state`](https://www.closefinch.com/docs/mcp/tools/ra_account_state) | Get an account's state: memory, each agent's decisions with evidence, and the next best action. | No | | [`ra_activities_create`](https://www.closefinch.com/docs/mcp/tools/ra_activities_create) | Log a call, meeting or note against records (it joins their timelines). | Yes | | [`ra_activities_get`](https://www.closefinch.com/docs/mcp/tools/ra_activities_get) | Read one email, meeting, call or note in full. | No | | [`ra_activities_list`](https://www.closefinch.com/docs/mcp/tools/ra_activities_list) | Find emails, meetings, calls and notes, for one record or a time range. | No | | [`ra_agents_add_trigger`](https://www.closefinch.com/docs/mcp/tools/ra_agents_add_trigger) | Add a trigger to an agent: a schedule (cron), an event or a webhook. | Yes | | [`ra_agents_create`](https://www.closefinch.com/docs/mcp/tools/ra_agents_create) | Create an agent as a draft, from a template or a spec. Nothing runs until it's published (ra_agents_publish). | Yes | | [`ra_agents_draft_from_goal`](https://www.closefinch.com/docs/mcp/tools/ra_agents_draft_from_goal) | Draft an agent from a goal in plain language (a draft to review, not a live agent). | Yes | | [`ra_agents_get`](https://www.closefinch.com/docs/mcp/tools/ra_agents_get) | Read one agent: its job, steps, triggers, permissions and version. | No | | [`ra_agents_list`](https://www.closefinch.com/docs/mcp/tools/ra_agents_list) | List the organization's agents and what each one does. | No | | [`ra_agents_publish`](https://www.closefinch.com/docs/mcp/tools/ra_agents_publish) | Publish an agent's draft as its new live version. | Yes | | [`ra_agents_test_run`](https://www.closefinch.com/docs/mcp/tools/ra_agents_test_run) | Test-run an agent's draft once, as a dry run (nothing is sent or changed). | Yes | | [`ra_agents_triggers`](https://www.closefinch.com/docs/mcp/tools/ra_agents_triggers) | List what starts an agent: schedules, events and webhooks. | No | | [`ra_agents_update`](https://www.closefinch.com/docs/mcp/tools/ra_agents_update) | Edit an agent's draft (the live version keeps running until you publish). | Yes | | [`ra_api`](https://www.closefinch.com/docs/mcp/tools/ra_api) | Call any Closefinch REST route (the spec is /v1/openapi.json) with your credentials. DELETE and merges need confirm: true. | Yes (asks for confirm) | | [`ra_approvals_get`](https://www.closefinch.com/docs/mcp/tools/ra_approvals_get) | Read one item waiting for approval: what would happen, why, and its evidence. | No | | [`ra_approvals_list`](https://www.closefinch.com/docs/mcp/tools/ra_approvals_list) | List what's waiting for a person's approval. Deciding happens in the app or Slack, never here. | No | | [`ra_ask`](https://www.closefinch.com/docs/mcp/tools/ra_ask) | Answer a question about your CRM in plain language, with the records it used as sources. | Yes | | [`ra_billing_upgrade_link`](https://www.closefinch.com/docs/mcp/tools/ra_billing_upgrade_link) | Suggest a paid plan to the organization's owners: get a link for an owner to review and pay. Nothing is bought here. | Yes | | [`ra_canvases_get`](https://www.closefinch.com/docs/mcp/tools/ra_canvases_get) | Read a canvas (a table or document a person or agent wrote). | No | | [`ra_conversations_ask`](https://www.closefinch.com/docs/mcp/tools/ra_conversations_ask) | Ask a follow-up question in an existing assistant conversation. | Yes | | [`ra_conversations_get`](https://www.closefinch.com/docs/mcp/tools/ra_conversations_get) | Read one assistant conversation with its messages and sources. | No | | [`ra_conversations_list`](https://www.closefinch.com/docs/mcp/tools/ra_conversations_list) | List your conversations with the assistant (questions asked in the app). | No | | [`ra_cpq_catalog`](https://www.closefinch.com/docs/mcp/tools/ra_cpq_catalog) | List the product catalog: products, prices and pricing rules for quotes. | No | | [`ra_evals_add`](https://www.closefinch.com/docs/mcp/tools/ra_evals_add) | Add an eval case to an agent: an input and what a good result looks like. | Yes | | [`ra_evals_list`](https://www.closefinch.com/docs/mcp/tools/ra_evals_list) | List an agent's eval cases (its regression tests). | No | | [`ra_evals_run`](https://www.closefinch.com/docs/mcp/tools/ra_evals_run) | Run an agent's eval suite as dry runs on its draft or live version and score them. | Yes | | [`ra_graph_neighborhood`](https://www.closefinch.com/docs/mcp/tools/ra_graph_neighborhood) | Map who's connected to a record: people, companies and deals around it, and how recently they talked. | No | | [`ra_home`](https://www.closefinch.com/docs/mcp/tools/ra_home) | Get today's summary: deals that need attention, meetings, waiting approvals and what agents did. | No | | [`ra_hygiene_issues`](https://www.closefinch.com/docs/mcp/tools/ra_hygiene_issues) | Find data problems: duplicates, missing fields and stale records, with suggested fixes. | No | | [`ra_kb_get`](https://www.closefinch.com/docs/mcp/tools/ra_kb_get) | Read one knowledge base section in full. | No | | [`ra_kb_list`](https://www.closefinch.com/docs/mcp/tools/ra_kb_list) | Read the company's knowledge base: positioning, products, proof points, voice, ICP and playbooks. | No | | [`ra_knowledge_search`](https://www.closefinch.com/docs/mcp/tools/ra_knowledge_search) | Search the company's documents (help center, decks, policies) and get passages with links. | No | | [`ra_list_records`](https://www.closefinch.com/docs/mcp/tools/ra_list_records) | Read the records in a saved list. | No | | [`ra_lists_list`](https://www.closefinch.com/docs/mcp/tools/ra_lists_list) | Find saved lists of records (segments), by object type. | No | | [`ra_memory_list`](https://www.closefinch.com/docs/mcp/tools/ra_memory_list) | Read what agents remember about a record (facts with their sources). | No | | [`ra_pipelines_list`](https://www.closefinch.com/docs/mcp/tools/ra_pipelines_list) | List the deal pipelines and their stages, with each stage's probability. | No | | [`ra_projects_get`](https://www.closefinch.com/docs/mcp/tools/ra_projects_get) | Read one project with its steps, owners and dates. | No | | [`ra_projects_list`](https://www.closefinch.com/docs/mcp/tools/ra_projects_list) | List projects (mutual action plans, onboarding plans). | No | | [`ra_propose`](https://www.closefinch.com/docs/mcp/tools/ra_propose) | Do something the way an agent would: reads now, sends and changes to existing records wait for a person's approval. | Yes | | [`ra_questionnaire_analytics`](https://www.closefinch.com/docs/mcp/tools/ra_questionnaire_analytics) | Measure questionnaire workload: volume, time to answer and who answers. | No | | [`ra_questionnaire_answer`](https://www.closefinch.com/docs/mcp/tools/ra_questionnaire_answer) | Answer a buyer's security or RFP questions from approved sources, each answer with citations. | Yes | | [`ra_quotes_create`](https://www.closefinch.com/docs/mcp/tools/ra_quotes_create) | Draft a quote on a deal from catalog products. Discounts beyond the rules wait for approval. | Yes | | [`ra_quotes_get`](https://www.closefinch.com/docs/mcp/tools/ra_quotes_get) | Read a quote with each line and number's evidence (catalog price, rule, discount). | No | | [`ra_quotes_intel`](https://www.closefinch.com/docs/mcp/tools/ra_quotes_intel) | Compare prices: what similar deals paid, discounts given and win rates by price. | No | | [`ra_quotes_list`](https://www.closefinch.com/docs/mcp/tools/ra_quotes_list) | Find quotes, for a deal, an owner or a status. | No | | [`ra_quotes_pipeline`](https://www.closefinch.com/docs/mcp/tools/ra_quotes_pipeline) | Find open quotes to chase next: oldest, biggest or about to expire. | No | | [`ra_records_archive`](https://www.closefinch.com/docs/mcp/tools/ra_records_archive) | Archive a record (it can be restored in the app). Needs confirm: true. | Yes (asks for confirm) | | [`ra_records_create`](https://www.closefinch.com/docs/mcp/tools/ra_records_create) | Create an account, contact, deal, lead or task. Check the fields with ra_records_schema first. | Yes | | [`ra_records_get`](https://www.closefinch.com/docs/mcp/tools/ra_records_get) | Read one account, contact, deal, lead or task with all its fields. | No | | [`ra_records_history`](https://www.closefinch.com/docs/mcp/tools/ra_records_history) | See who changed a record's fields, when, and from what (people, agents, sync). | No | | [`ra_records_list`](https://www.closefinch.com/docs/mcp/tools/ra_records_list) | Find accounts, contacts, deals, leads or tasks: filter by name (q), owner or any field (filter[field]=value), newest first. | No | | [`ra_records_schema`](https://www.closefinch.com/docs/mcp/tools/ra_records_schema) | List the fields an object has (built-in and custom), with their types and allowed values. | No | | [`ra_records_timeline`](https://www.closefinch.com/docs/mcp/tools/ra_records_timeline) | Read a record's timeline: emails, meetings, calls and notes, newest first, each with its source. | No | | [`ra_records_update`](https://www.closefinch.com/docs/mcp/tools/ra_records_update) | Change fields on a record; pass version to avoid overwriting someone else's edit. | Yes | | [`ra_runs_cancel`](https://www.closefinch.com/docs/mcp/tools/ra_runs_cancel) | Cancel a running agent run. | Yes | | [`ra_runs_get`](https://www.closefinch.com/docs/mcp/tools/ra_runs_get) | Read an agent run's trace: each step, its tool calls, cost and anything waiting for approval. | No | | [`ra_runs_list`](https://www.closefinch.com/docs/mcp/tools/ra_runs_list) | List agent runs, newest first, by agent or status. | No | | [`ra_runs_start`](https://www.closefinch.com/docs/mcp/tools/ra_runs_start) | Start an agent run now (dry_run: true to see what it would do without doing it). | Yes | | [`ra_telemetry`](https://www.closefinch.com/docs/mcp/tools/ra_telemetry) | Measure how agents are doing: runs, success rate, cost, approvals and time saved, by agent. | No | | [`ra_templates_list`](https://www.closefinch.com/docs/mcp/tools/ra_templates_list) | Browse agent templates (built-in and your team's) to start an agent from. | No | | [`ra_tools_list`](https://www.closefinch.com/docs/mcp/tools/ra_tools_list) | List the platform tools agents can use (for ra_propose), with what each reads or changes. | No | | [`ra_whoami`](https://www.closefinch.com/docs/mcp/tools/ra_whoami) | Check who you're signed in as: your organization, role and scopes. Call it first to see what you may do. | No | Connecting a client: [quickstarts](https://www.closefinch.com/docs/quickstart). What runs at once and what waits for a person: [approvals and policy](https://www.closefinch.com/docs/approvals). --- Source: https://www.closefinch.com/docs/api # API reference The REST API at `https://api.closefinch.com/v1`, by resource. The full spec, with request and response schemas, is [https://www.closefinch.com/openapi.json](https://www.closefinch.com/openapi.json) (OpenAPI 3.1). - [Account](https://www.closefinch.com/docs/api/account): 15 routes - [Account plans](https://www.closefinch.com/docs/api/account-plans): 12 routes - [Accounts](https://www.closefinch.com/docs/api/accounts): 20 routes - [Activity](https://www.closefinch.com/docs/api/activity): 12 routes - [Agent browser](https://www.closefinch.com/docs/api/agent-browser): 13 routes - [Agent mailboxes](https://www.closefinch.com/docs/api/agent-mailboxes): 8 routes - [Agent platform](https://www.closefinch.com/docs/api/agent-platform): 78 routes - [Agents](https://www.closefinch.com/docs/api/agents): 9 routes - [AI fields](https://www.closefinch.com/docs/api/ai-fields): 15 routes - [Analytics](https://www.closefinch.com/docs/api/analytics): 26 routes - [API keys](https://www.closefinch.com/docs/api/api-keys): 5 routes - [Ask](https://www.closefinch.com/docs/api/ask): 4 routes - [Audit log](https://www.closefinch.com/docs/api/audit-log): 4 routes - [Billing](https://www.closefinch.com/docs/api/billing): 10 routes - [Browser extension](https://www.closefinch.com/docs/api/browser-extension): 5 routes - [Business profile](https://www.closefinch.com/docs/api/business-profile): 10 routes - [Calling compliance](https://www.closefinch.com/docs/api/calling-compliance): 26 routes - [Catalog](https://www.closefinch.com/docs/api/catalog): 3 routes - [Company signals](https://www.closefinch.com/docs/api/company-signals): 4 routes - [Competitive intelligence](https://www.closefinch.com/docs/api/competitive-intelligence): 17 routes - [Contacts](https://www.closefinch.com/docs/api/contacts): 17 routes - [Conversations](https://www.closefinch.com/docs/api/conversations): 9 routes - [CRM migration](https://www.closefinch.com/docs/api/crm-migration): 8 routes - [CRM sync](https://www.closefinch.com/docs/api/crm-sync): 10 routes - [Custom object records](https://www.closefinch.com/docs/api/custom-object-records): 17 routes - [Customers](https://www.closefinch.com/docs/api/customers): 49 routes - [Data hygiene](https://www.closefinch.com/docs/api/data-hygiene): 7 routes - [Deal intelligence](https://www.closefinch.com/docs/api/deal-intelligence): 42 routes - [Deals](https://www.closefinch.com/docs/api/deals): 17 routes - [Deliverability](https://www.closefinch.com/docs/api/deliverability): 14 routes - [Desktop](https://www.closefinch.com/docs/api/desktop): 10 routes - [Dialer](https://www.closefinch.com/docs/api/dialer): 15 routes - [Embeds](https://www.closefinch.com/docs/api/embeds): 3 routes - [Emergency dispatch](https://www.closefinch.com/docs/api/emergency-dispatch): 10 routes - [Engagement and intent](https://www.closefinch.com/docs/api/engagement-and-intent): 10 routes - [Enrichment](https://www.closefinch.com/docs/api/enrichment): 31 routes - [Entity mappings](https://www.closefinch.com/docs/api/entity-mappings): 4 routes - [Feedback](https://www.closefinch.com/docs/api/feedback): 3 routes - [First-party signals](https://www.closefinch.com/docs/api/first-party-signals): 22 routes - [Forecasting](https://www.closefinch.com/docs/api/forecasting): 34 routes - [Freshness](https://www.closefinch.com/docs/api/freshness): 18 routes - [Glossary](https://www.closefinch.com/docs/api/glossary): 7 routes - [Governance](https://www.closefinch.com/docs/api/governance): 5 routes - [Imports](https://www.closefinch.com/docs/api/imports): 5 routes - [Integrations](https://www.closefinch.com/docs/api/integrations): 17 routes - [Interview synthesis](https://www.closefinch.com/docs/api/interview-synthesis): 5 routes - [Knowledge](https://www.closefinch.com/docs/api/knowledge): 7 routes - [Knowledge base](https://www.closefinch.com/docs/api/knowledge-base): 7 routes - [Lead routing](https://www.closefinch.com/docs/api/lead-routing): 28 routes - [Leads](https://www.closefinch.com/docs/api/leads): 18 routes - [Links](https://www.closefinch.com/docs/api/links): 1 route - [Lists](https://www.closefinch.com/docs/api/lists): 7 routes - [Live](https://www.closefinch.com/docs/api/live): 8 routes - [Market](https://www.closefinch.com/docs/api/market): 14 routes - [Marketing](https://www.closefinch.com/docs/api/marketing): 36 routes - [MCP servers](https://www.closefinch.com/docs/api/mcp-servers): 5 routes - [Me](https://www.closefinch.com/docs/api/me): 2 routes - [Meetings](https://www.closefinch.com/docs/api/meetings): 40 routes - [Memory and graph](https://www.closefinch.com/docs/api/memory-and-graph): 9 routes - [Messaging](https://www.closefinch.com/docs/api/messaging): 28 routes - [Meta](https://www.closefinch.com/docs/api/meta): 2 routes - [Microsoft Teams](https://www.closefinch.com/docs/api/microsoft-teams): 3 routes - [Notes](https://www.closefinch.com/docs/api/notes): 17 routes - [Notifications](https://www.closefinch.com/docs/api/notifications): 9 routes - [OAuth apps](https://www.closefinch.com/docs/api/oauth-apps): 12 routes - [Organization](https://www.closefinch.com/docs/api/organization): 17 routes - [Outbound](https://www.closefinch.com/docs/api/outbound): 59 routes - [Owners](https://www.closefinch.com/docs/api/owners): 1 route - [Packs](https://www.closefinch.com/docs/api/packs): 21 routes - [Packs: CPG](https://www.closefinch.com/docs/api/packs-cpg): 28 routes - [Packs: equipment](https://www.closefinch.com/docs/api/packs-equipment): 24 routes - [Packs: fundraising](https://www.closefinch.com/docs/api/packs-fundraising): 27 routes - [Packs: home services](https://www.closefinch.com/docs/api/packs-home-services): 31 routes - [Partners and investors](https://www.closefinch.com/docs/api/partners-and-investors): 8 routes - [Payments](https://www.closefinch.com/docs/api/payments): 6 routes - [People data](https://www.closefinch.com/docs/api/people-data): 15 routes - [Phone and SMS](https://www.closefinch.com/docs/api/phone-and-sms): 21 routes - [Phone menus and transfers](https://www.closefinch.com/docs/api/phone-menus-and-transfers): 20 routes - [Phone numbers and calls](https://www.closefinch.com/docs/api/phone-numbers-and-calls): 30 routes - [Pipelines](https://www.closefinch.com/docs/api/pipelines): 5 routes - [Plays](https://www.closefinch.com/docs/api/plays): 27 routes - [Prioritization](https://www.closefinch.com/docs/api/prioritization): 18 routes - [Projects](https://www.closefinch.com/docs/api/projects): 10 routes - [Proof points](https://www.closefinch.com/docs/api/proof-points): 5 routes - [Prospecting](https://www.closefinch.com/docs/api/prospecting): 53 routes - [Public records](https://www.closefinch.com/docs/api/public-records): 6 routes - [Questionnaires](https://www.closefinch.com/docs/api/questionnaires): 31 routes - [Quotes](https://www.closefinch.com/docs/api/quotes): 64 routes - [Recording](https://www.closefinch.com/docs/api/recording): 22 routes - [Records](https://www.closefinch.com/docs/api/records): 1 route - [Replies](https://www.closefinch.com/docs/api/replies): 30 routes - [Research](https://www.closefinch.com/docs/api/research): 8 routes - [Sales floor](https://www.closefinch.com/docs/api/sales-floor): 15 routes - [Sandbox](https://www.closefinch.com/docs/api/sandbox): 4 routes - [Sandboxes](https://www.closefinch.com/docs/api/sandboxes): 4 routes - [Scheduling](https://www.closefinch.com/docs/api/scheduling): 20 routes - [Schema](https://www.closefinch.com/docs/api/schema): 14 routes - [Scraping](https://www.closefinch.com/docs/api/scraping): 14 routes - [Settings](https://www.closefinch.com/docs/api/settings): 10 routes - [Shared background](https://www.closefinch.com/docs/api/shared-background): 3 routes - [Signals](https://www.closefinch.com/docs/api/signals): 27 routes - [Single sign-on](https://www.closefinch.com/docs/api/single-sign-on): 10 routes - [Social signals](https://www.closefinch.com/docs/api/social-signals): 24 routes - [Tasks](https://www.closefinch.com/docs/api/tasks): 19 routes - [Vendors](https://www.closefinch.com/docs/api/vendors): 32 routes - [Voice](https://www.closefinch.com/docs/api/voice): 9 routes - [Voice agents](https://www.closefinch.com/docs/api/voice-agents): 19 routes - [Voice calls](https://www.closefinch.com/docs/api/voice-calls): 9 routes - [Voice campaigns](https://www.closefinch.com/docs/api/voice-campaigns): 11 routes - [Website chat](https://www.closefinch.com/docs/api/website-chat): 26 routes - [Workspaces](https://www.closefinch.com/docs/api/workspaces): 6 routes Every request sends `Authorization: Bearer ra_live_…`. Errors, rate limits and what to do about each: [limits and errors](https://www.closefinch.com/for-agents#limits-and-errors). Versions and deprecations: [changelog](https://www.closefinch.com/docs/changelog). --- Source: https://www.closefinch.com/terms # Terms of service Effective November 8, 2026. Last updated: November 8, 2026. These terms are an agreement between Ask Lou LLC, a Delaware limited liability company at 160 Maplewood Ave, Maplewood, NJ 07040, USA ("Closefinch", "we"), and the business that creates or uses an account ("Customer", "you"). By creating an account or using the service, you agree to them on behalf of that business, and you confirm you can bind it. The service is for business use only, not for consumers, and is offered to businesses in the United States. ## 1. The service Closefinch is a hosted CRM and agent platform. It: - captures email, calendar and meeting activity you connect; - keeps records of accounts, people and deals; - runs AI features and agents you configure; - offers an API, an MCP server and a CLI. Features differ by plan, as the [pricing page](https://www.closefinch.com/pricing) describes. We may improve and change the service, and we won't materially reduce a paid plan's core features during a term you've paid for. ## 2. Accounts and security - **Keep credentials safe.** Use two-step sign-in, which is required for owners. Tell us at once about any unauthorized use. - **Your people.** You're responsible for the people you invite, and for API keys and apps you authorize. - **Our protective measures.** We may suspend access, or hold outbound sending, to protect the service, other customers or the people you contact. Examples: suspected account takeover, spam, a security incident, or a breach of these terms. We'll tell you why, unless the law or an investigation prevents it, and restore access once the issue is resolved. ## 3. Your data - **"Customer Data"** is everything you or your connected accounts put into the service, including the personal data of people in your emails, meetings and records. - **It's yours.** You give us the rights needed to host it, process it and show it back to you, only to provide and secure the service and as you instruct. Our [Data Processing Agreement](https://www.closefinch.com/dpa) is part of these terms and governs personal data. - **No training, no selling.** We don't sell Customer Data, and we don't use it to train AI models. - **Usage data.** We may use aggregated, de-identified data about how the service is used (counts and timings, never content) to run and improve it. - **Export and deletion.** You can export your data at any time. After you close your account we delete Customer Data within 30 days (backups expire on their schedule), unless the law requires us to keep it. ## 4. Your responsibilities You're responsible for having the rights and lawful basis to put Customer Data into the service, and to have it act for you. In particular: - **Recording.** You choose whether to record meetings. You're responsible for notice and for consent where the law requires it, including states and countries that require every party to agree. The service names the recorder, posts a notice and lets anyone stop it, but that doesn't replace your obligations. - **Email and messages.** Messages the service sends for you go from your own mailbox or accounts and are your messages. You must follow anti-spam and marketing law, including CAN-SPAM, PECR and ePrivacy, CASL and consent rules, and the [Acceptable Use Policy](https://www.closefinch.com/acceptable-use). - **AI output.** AI features can be wrong or incomplete. Review what matters before you rely on it or send it. Outgoing actions wait for a person's approval by default, and loosening that is your choice. - **Connected services.** Your use of Google, Microsoft, Slack, CRMs and other services you connect is governed by your agreements with them. ## 5. Acceptable use You'll follow the [Acceptable Use Policy](https://www.closefinch.com/acceptable-use). Under it, the service can't be used for spam, deception, unlawful surveillance, discrimination, or attacks on the service or others. ## 6. Fees - **Paid plans** are billed in advance by Paddle, our merchant of record, which also handles sales tax and VAT. - **Usage beyond a plan's allowance** is billed at the rates on the pricing page, such as recording hours or AI credits. - **Price changes** for existing subscriptions take effect at renewal, after at least 30 days' notice. - **Refunds.** Fees are non-refundable except where the law or these terms say otherwise. - **Trials and free plans** have limits and may change or end. ## 7. Confidentiality Each party keeps the other's non-public information confidential, uses it only for this agreement, and protects it with at least reasonable care. ## 8. Warranties and disclaimers - **What we warrant:** we'll provide the service with reasonable skill and care, consistent with our security overview. - **Everything else is excluded.** Otherwise the service is provided "as is". To the extent the law allows, we disclaim implied warranties of merchantability, fitness for a particular purpose and non-infringement, and any warranty that AI output is accurate. ## 9. Liability - **Excluded losses.** To the extent the law allows, neither party is liable for indirect, incidental, special, consequential or punitive damages, or for lost profits, revenue or data. Data is excepted where the loss is caused by a breach of our security obligations. - **Cap.** Each party's total liability is capped at the fees you paid in the 12 months before the claim, or USD 100 if you're on a free plan. - **Exceptions to the cap:** your payment obligations, breach of section 4 or the Acceptable Use Policy, indemnities, and liability that can't be limited by law. ## 10. Indemnity You'll defend us against third-party claims arising from Customer Data you weren't entitled to use, or from your messages, recordings or breach of the Acceptable Use Policy. We'll defend you against third-party claims that the service, used as permitted, infringes their intellectual property. ## 11. Term and termination - **Term.** These terms last while you have an account. - **You can cancel at any time.** A paid plan runs to the end of its term. - **Termination for breach.** Either party may end the agreement for a material breach not cured within 30 days of notice. We may end it at once for serious abuse. - **What survives:** sections 3 (deletion), 7, 8, 9, 10 and 13. ## 12. Changes to these terms We'll post changes here, and email owners at least 30 days before a material change takes effect. If you don't agree, you can cancel before then. ## 13. General - **Governing law and courts.** These terms are governed by the laws of the State of Delaware, without regard to its conflict-of-laws rules. The state and federal courts in Delaware have exclusive jurisdiction. - **Assignment.** Neither party may assign this agreement without the other's consent, except to a successor in a merger or sale. - **Force majeure.** Neither party is liable for delays caused by events beyond its reasonable control. - **Entire agreement.** These terms, the DPA, the Acceptable Use Policy and any order form are the whole agreement. If they conflict, the DPA wins for personal data. - **Notices.** We send notices to account owners' emails. You send them to legal@closefinch.com and Ask Lou LLC, 160 Maplewood Ave, Maplewood, NJ 07040, USA. --- Source: https://www.closefinch.com/privacy # Privacy policy Effective November 8, 2026. Last updated: November 8, 2026. This policy explains how Ask Lou LLC, a Delaware limited liability company at 160 Maplewood Ave, Maplewood, NJ 07040, USA ("Closefinch", "we", "us"), handles personal data. It covers our website and the Closefinch service. ## 1. Two roles - **For our customers' data we're a processor (a "service provider" under US state law).** A company that signs up (the "customer") decides what to connect and what Closefinch does with it: email, calendars, meeting recordings, CRM records, and the people in them. We process that data only on the customer's instructions, under our [Data Processing Agreement](https://www.closefinch.com/dpa). If you're a person in a customer's emails or meetings and want to use your privacy rights, contact that company. We'll help them answer you. - **For our own users, visitors and buyers we're a controller.** That covers account details, billing contacts, website visits and support conversations. The rest of this policy covers that data, and how we treat customer data in practice. ## 2. What we collect, and why | Data | Where it comes from | Why (and our legal basis under GDPR) | |---|---|---| | Account: name, email, password hash, two-step sign-in settings, organization and role | You, at sign-up | To provide the service you signed up for (contract) and keep it secure (legitimate interests) | | Security records: sign-in events, IP address, device and browser, security alerts | Our systems | To protect accounts and the service, detect abuse and investigate incidents (legitimate interests; legal obligation where it applies) | | Billing: billing contact, plan, invoices | You, via Paddle (our merchant of record; Paddle holds payment details) | To charge for the service and meet tax and accounting duties (contract; legal obligation) | | Product use: which features an organization uses, counts and timings | Our systems | To run, secure and improve the service, and to see where people get stuck (legitimate interests). No advertising profiles. | | Session replay (only when an owner of your organization turns it on; each person can opt out): which app pages are opened, the shape of each page with every letter and number masked (the length of words stays visible), clicks, scrolling and window size. Never what you type, and never sign-in, account, key, secret, billing or payment pages, or a page while it shows a password field or a secret | Your browser | To find screens that confuse people and fix them (legitimate interests) | | Website: pages visited, referrer, approximate country | Our cookieless web counter | To understand how people find us (legitimate interests). No tracking cookies and no third-party analytics scripts. | | Support and sales conversations | You | To answer you (contract; legitimate interests) | ## 3. Customer data in practice - **Connected accounts.** When someone in a customer's organization connects Gmail, Outlook, a calendar or a CRM, we read only what the features they turn on need. Tokens are encrypted. People can disconnect at any time, and what was captured through a connection can be removed with it. - **Meeting recordings.** The recorder joins only when the organization has turned it on and the person who scheduled the meeting has agreed to it. The bot's name says it records, and it posts a notice with a link that anyone can use to stop it. Stopping deletes the media. The customer is responsible for having a lawful basis and for consent where the law requires all parties to agree. Media is deleted after 7 days (30 for meetings with clips). Transcripts are kept for 2 years by default, and each organization can set its own periods. - **AI.** Features that use AI send the text a task needs to our AI provider (see [Subprocessors](https://www.closefinch.com/subprocessors)). Neither we nor our providers use customer data to train AI models. AI output can be wrong: the product shows its sources, and outgoing messages wait for a person's approval by default. - **Feedback and error reports.** What you send us as feedback (from the app or the website), and the error reports the service records, may be processed by AI tools our team uses to diagnose and fix problems. Email addresses, phone numbers, card numbers and secrets are masked before they do, and a person on our team approves any reply to you. An email you give with feedback from the website is used only to reply about it, never for marketing. **[COUNSEL REVIEW: added for the ops inbox, OPS-AUTOFIX]** - **Google user data.** Closefinch's use and transfer of information received from Google APIs adheres to the [Google API Services User Data Policy](https://developers.google.com/terms/api-services-user-data-policy), including its **Limited Use** requirements. We use Gmail and Calendar data only to provide the features the user turns on. We don't use it for advertising and don't sell it. We don't let people read it except: with the user's agreement; for security (such as investigating abuse); to comply with law; or when it's aggregated and anonymized for operations. We don't use it to train AI models. ## 4. Who we share data with - **Our [subprocessors](https://www.closefinch.com/subprocessors),** under contract, for the purposes above. - **Authorities,** when the law requires it. We push back on requests that are overbroad and, where we're allowed to, tell the customer first. - **A successor,** if Ask Lou LLC is acquired or merged. This policy keeps applying to the data. We **don't sell** personal data and **don't share** it for cross-context behavioral advertising. ## 5. International transfers We host in Germany. Some subprocessors are in the United States. Transfers out of the EU, EEA and UK are covered by the European Commission's Standard Contractual Clauses and the UK Addendum, and by the EU-US Data Privacy Framework where a provider is certified. ## 6. How long we keep data - **Account data** while the account exists, then up to 30 days in backups. - **Security records** for 1 year. - **Billing records** as long as tax law requires (typically 7 years). - **Feedback** you send from the app (and survey answers) for 2 years, or until your account or organization is deleted. - **Session replay recordings** for 30 days, or until you opt out, an owner turns it off, or your account or organization is deleted. - **Customer data** as the customer's settings and our DPA say. When an organization is deleted, its data is purged, including its audit log, and backups expire within 30 days. ## 7. Security Safeguards include: - separation of each organization's data, enforced by the database; - encryption in transit and at rest (the database disk and every backup), and encrypted connected-account tokens; - two-step sign-in, which owners need to use owner powers; - a tamper-evident audit log; - automated detection and alerting; - the least access our own staff need. More is in our [security overview](https://www.closefinch.com/security). ## 8. Your rights Depending on where you live (for example under the GDPR, UK GDPR or California's CCPA/CPRA), you may have the right to: - access your personal data, correct it, delete it, or get a copy of it; - restrict or object to our processing; - withdraw consent; - not be discriminated against for using these rights. Owners can ask us for a full export of their organization's data, and users can delete their account in the product. Otherwise, email privacy@closefinch.com. We answer within 30 days (45 under CCPA), and we may need to confirm who you are. You may also complain to your data protection authority. We offer the service to businesses in the United States and don't direct it to people in the EU or the UK, so we haven't appointed representatives there. ## 9. Children The service is for businesses and isn't directed to children under 16. We don't knowingly collect their data. ## 10. Changes We'll post changes here, and tell account owners by email before a material change takes effect. ## 11. Contact Ask Lou LLC, 160 Maplewood Ave, Maplewood, NJ 07040, USA · privacy@closefinch.com --- Source: https://www.closefinch.com/dpa # Data processing agreement Effective November 8, 2026. Last updated: November 8, 2026. This Data Processing Agreement ("DPA") is part of the [Terms of Service](https://www.closefinch.com/terms) between Ask Lou LLC, a Delaware limited liability company at 160 Maplewood Ave, Maplewood, NJ 07040, USA ("Processor", "we"), and the Customer ("Controller", "you"). It applies whenever we process Customer Personal Data for you. It is meant to meet Article 28 of the GDPR and UK GDPR, and the service-provider requirements of the CCPA/CPRA and similar US state laws. ## 1. Definitions Terms used in the GDPR (such as personal data, processing, controller, processor, data subject and personal data breach) have the meanings given there. "**Customer Personal Data**" means personal data in Customer Data that we process for you. "**Subprocessor**" means a third party we engage to process Customer Personal Data. ## 2. Roles and instructions You're the controller (or a processor acting for your controller), and we're your processor. We process Customer Personal Data only on your documented instructions. Those are: these terms; your configuration of the service (connections, features, agents, retention settings); and your other written instructions. We'll tell you if we believe an instruction breaks data-protection law. As a CCPA service provider we won't: - sell or share Customer Personal Data; - retain, use or disclose it outside our direct business relationship with you, or for any purpose other than providing the service; - combine it with other data except as the law allows. ## 3. Details of the processing (Annex I) | | | |---|---| | Subject matter and duration | Providing the service, for the term of the agreement and until deletion under section 10 | | Nature and purpose | Hosting, capturing, organizing, searching, summarizing and acting on Customer Data as you configure: email and calendar capture, CRM records, meeting recording and transcription, AI features and agents, the API, MCP and integrations | | Categories of data subjects | Your users; your contacts, prospects, customers and partners; meeting participants; anyone else in Customer Data | | Categories of personal data | Names, business contact details, job details, email and message content and metadata, calendar entries, meeting audio, video and transcripts, CRM notes and fields, and whatever else you put in | | Special categories | Not intended. You won't submit them unless needed and lawful. The service redacts some sensitive identifiers automatically | | Frequency | Continuous | | Retention | As configured by you (defaults: recording media 7 days, or 30 with clips; transcripts 2 years), then as in section 10 | ## 4. Confidentiality and personnel Everyone we authorize to process Customer Personal Data is bound by confidentiality. Our staff access Customer Data only as needed to provide support you've asked for, keep the service secure, or comply with law, and that access is logged. ## 5. Security (Annex II) We maintain the technical and organizational measures described in our [security overview](https://www.closefinch.com/security). They include: - separation of each organization's data, enforced by the database; - encryption in transit (TLS) and at rest (the database disk with LUKS2, AES-256; backups with OpenPGP before they leave the server, to a key kept off our servers), with connected-account credentials encrypted by us under versioned keys that can be rotated; - two-step sign-in, which owners need to use owner powers; - least-privilege access, for our own staff too; - an append-only, tamper-evident audit log; - automated detection, alerting, the means to stop sending at once, and incident response; - encrypted backups, with restore drills on a schedule; - dependency controls; - secure development with independent review of every release. We may update these measures, but never in a way that lowers overall protection. ## 6. Subprocessors You give general authorization for the subprocessors listed at [Subprocessors](https://www.closefinch.com/subprocessors). - **Notice.** We give at least 30 days' notice of a new or replacement subprocessor, by email and on that page. - **Objections.** You may object on reasonable data-protection grounds within that period. We'll then try to find a solution. If we can't, you may end the affected service and get a pro-rata refund of prepaid fees for it. - **Their terms.** We impose data-protection terms on each subprocessor at least as protective as this DPA, and we're responsible for their performance. ## 7. Data subject requests and assistance The service lets you access, correct, export and delete Customer Personal Data. If a data subject contacts us directly, we'll pass the request to you and won't answer it ourselves unless you tell us to. We'll help you, taking into account the nature of the processing, with: - data subject requests; - security; - breach notification; - data protection impact assessments; - prior consultation with regulators. ## 8. Personal data breaches We'll notify you without undue delay after becoming aware of a personal data breach affecting Customer Personal Data, and in any case within **48 hours**. The notice will say what we know: the nature of the breach, the categories and approximate numbers involved, the likely consequences, and the measures taken or proposed. We'll follow up as we learn more. Notifying us of a breach isn't an admission of fault. ## 9. International transfers We host Customer Data in Germany. Where Customer Personal Data subject to the GDPR or UK GDPR is transferred to a country without an adequacy decision, the following apply, with Annexes I and II of this DPA and the subprocessor list as their annexes: - the **Standard Contractual Clauses** (Commission Implementing Decision (EU) 2021/914), Module 2 (controller-to-processor) or Module 3 (processor-to-processor); - the **UK International Data Transfer Addendum**. The governing law and forum for the Clauses are Ireland. Docking (Clause 7) applies. For Clause 9, the option is general authorization with the notice period above. We'll also rely on the EU-US Data Privacy Framework where a recipient is certified. ## 10. Deletion and return You can export Customer Data at any time while the agreement is in force. After it ends, we delete Customer Personal Data within 30 days, and backups expire on their schedule (within 30 more days), unless the law requires us to keep it. In that case we keep it confidential and process it only for that purpose. ## 11. Audits We'll make available the information needed to show compliance with this DPA: our security overview, questionnaire answers, and independent audit reports once we have them. If that isn't enough, you may audit us, or have an independent auditor do so. That's limited to once a year (more often after a breach or where a regulator requires it), with 30 days' notice, during business hours, under confidentiality and at your cost. ## 12. Liability and precedence Liability under this DPA is subject to the limits in the terms, except where the law forbids limiting it. If this DPA conflicts with the terms, this DPA wins for personal data. If the Clauses conflict with this DPA, the Clauses win. --- Representatives: none appointed (the service is offered to businesses in the United States). Contact for this DPA: privacy@closefinch.com. --- Source: https://www.closefinch.com/subprocessors # Subprocessors Last updated: November 8, 2026. Ask Lou LLC ("Closefinch", "we") uses these companies to provide the service. Each one processes Customer Personal Data only as the service needs. Each is bound by written terms that protect the data at least as well as our Data Processing Agreement. Each is covered by Standard Contractual Clauses (or the EU-US Data Privacy Framework) wherever data leaves the EU or UK. We give at least **30 days' notice** before adding or replacing a subprocessor: by email to each customer's owners, and on this page. A customer may object for reasonable data-protection reasons, as the DPA sets out. Last updated: November 8, 2026 | Subprocessor | What it does for us | Data it processes | Location | When it's used | |---|---|---|---|---| | Hetzner Online GmbH | Hosting: servers, database, backups | All Customer Data (encrypted at rest: the database disk with LUKS2, backups with OpenPGP; connected-account tokens additionally encrypted by us) | Germany | Always | | Cloudflare, Inc. | Network edge for the API: TLS, DDoS protection, web application firewall | Traffic to the API in transit, including the calls the web app forwards | Global edge network | Always | | Vercel Inc. | Hosting the web app and the marketing site | Traffic to the site and the app in transit (page requests, and the API calls the app forwards); no database access | Global edge network; the app's functions run in Germany | Always | | Anthropic, PBC | AI models (Claude) for summaries, drafting, field filling, agents and answers; and our own operator support tooling: reading redacted feedback and error reports to diagnose and fix problems **[COUNSEL REVIEW: added for the ops inbox, OPS-AUTOFIX]** | The text the task needs (for example an email thread, a transcript, record fields). For support tooling: the feedback text and error reports, with email addresses, phone numbers, card numbers and secrets masked first; the reporter is never named | United States | When an AI feature runs. Not used for model training (commercial terms) and kept only as Anthropic's API terms allow. An organization can bring its own Anthropic key, in which case its own agreement with Anthropic applies. Support tooling: when one of our operators diagnoses a reported problem | | Voyage AI, Inc. | Text embeddings for semantic search | Knowledge base and document text | United States | Only when semantic search is on (Settings → AI; owners can turn it off, which deletes stored vectors) | | Recall.ai, Inc. | Meeting recorder bot | Meeting audio, video and participant names during the meeting; media kept 7 days (30 for meetings with clips) | United States (EU region for EU deployments) | Only when an organization turns on the recorder and a person has consented to its use | | AssemblyAI, Inc. | Transcription of uploaded recordings and in-person audio | Audio; transcripts deleted at AssemblyAI once we've copied them | United States (EU endpoint for EU deployments) | Only when a recording is uploaded or captured in person | | Daytona Platforms, Inc. | Sandbox for agent code steps | The inputs an organization's own published code step receives | United States / EU | Only when an organization publishes an agent with a code step | | Resend (Plus Five Five, Inc.) | Sending service email (sign-in links, invitations, digests, alerts) | Recipient address, message content | United States | Always | | Postmark (ActiveCampaign, LLC) | Receiving email sent to agent inboxes | Inbound messages to an agent's address | United States | Only when an organization uses an agent inbox | | ElevenLabs, Inc. | Text-to-speech for an AI phone agent's voice and persona previews | The text the agent speaks | United States | Only when an organization off the trial picks an ElevenLabs voice without bringing its own key | | Cartesia, Inc. | Text-to-speech (an alternative) | The text the agent speaks | United States | As for ElevenLabs, with a Cartesia voice | | Paddle.com Market Limited | Merchant of record: checkout, invoicing, sales tax and VAT | Billing contact, payment details (held by Paddle, never by us), plan and usage totals | United Kingdom / EU | When an organization buys a paid plan | ## Providers you connect with your own account These run only when an organization connects its own account, with its own key. They process what the organization sends them on its instructions, under its own agreement with the provider. They're listed so you know where that data goes. | Provider | What it does | Data it receives | When it's used | |---|---|---|---| | ElevenLabs, Inc. | Text-to-speech and voice cloning | The text to speak; for a clone, the teammate's own consent recording | Voice clones, AI voice notes in a teammate's own voice, and hold and voicemail clips: only with the organization's own ElevenLabs key | | Cartesia, Inc. | Text-to-speech and voice cloning (an alternative) | As for ElevenLabs | As for ElevenLabs, with the organization's own Cartesia key | | HeyGen (HeyGen Technology Inc.) | AI video of a teammate from their own photo | The teammate's own photo; the audio in their own cloned voice, which HeyGen fetches and keeps (the script may name a prospect's first name and company, and any other words the teammate writes); HeyGen keeps the audio and the rendered video until they're deleted, which happens when consent is withdrawn | Only with the organization's own HeyGen account, after the teammate's written consent | | Tavus, Inc. | AI video from a replica trained on a teammate's own video | The training video and the script (which may name a prospect's first name and company, and any other words the teammate writes); Tavus keeps the replica and videos until they're deleted, which happens when consent is withdrawn | Only with the organization's own Tavus account, after the teammate's written consent | | ScreenshotOne | Screenshots of public web pages | The page address | Only with the organization's own ScreenshotOne account | | AgentMail, Inc. | Agent mailboxes: sending and receiving email for an organization's AI agents at their own addresses | Agent mailbox messages (recipient and sender addresses, message content), bounce and complaint reports | Only with the organization's own AgentMail account (off by default; United States, or the EU region when the organization chooses it). Sign-in, invitation and notification email never goes through it | | Telnyx, Inc. | Calls and texts, and browser audio for the dialer | Call audio, phone numbers, message text | Only with the organization's own Telnyx account | The platform's own ElevenLabs or Cartesia key is used only for an AI phone agent's voice and persona previews, and only when the organization hasn't brought its own. That case is listed in the table above. Services that act on a customer's own instructions with the customer's own accounts aren't our subprocessors. The customer's agreements with those providers apply, for example Google Workspace, Microsoft 365, Slack, Zoom, HubSpot or Salesforce when the customer connects them. --- Source: https://www.closefinch.com/acceptable-use # Acceptable use policy Effective November 8, 2026. Last updated: November 8, 2026. This policy applies to everyone who uses Closefinch, including the agents you set up. If you break it, we may hold your outbound sending, suspend your organization or close your account (see the [terms](https://www.closefinch.com/terms), section 2). ## Email and outreach - **No spam.** Don't send bulk or unsolicited messages where the law requires consent, or to purchased or scraped lists that the law doesn't allow you to email. - **Honour opt-outs.** Act on them promptly. CAN-SPAM requires this within 10 business days, and the service does it automatically once outbound features are on. - **Don't deceive.** No misleading sender names, subjects or impersonation. Don't hide that a message was written by AI when the law requires you to disclose it. - **Respect send limits.** Don't try to get around them by creating more accounts or organizations. ## Recording and monitoring - Record meetings only with the notice and consent the law requires, including all-party consent where that applies. Don't remove or hide the recorder's notice. - Don't use the service to monitor people secretly, or to profile them in ways the law forbids. ## Data - Put only data into the service that you're allowed to. - Don't put in special-category data you don't need: health, biometric, financial account or government ID numbers. The service redacts some of it automatically, but that doesn't make it allowed. - Don't use the service to make decisions about people's employment, credit, housing or insurance. ## The service itself - **No probing or attacks.** Don't try to access other customers' data, get around security or limits, or probe or scan the service. Report vulnerabilities under our [disclosure policy](https://www.closefinch.com/security#reporting-a-vulnerability) instead. - **No overloading.** Don't overload the service, resell it without our agreement, or use it to build a competing product from its outputs. - **Agents.** Don't use agents, the API or the MCP server to do anything this policy forbids. You're responsible for what agents you configure do. ## Reporting abuse Write to abuse@closefinch.com with what happened and any evidence. --- Source: https://www.closefinch.com/security # Security at Closefinch Last updated: November 8, 2026. Closefinch holds teams' email, calendars, CRM records and recordings, and can send messages for them. These are the controls that protect that, and the ones you can turn on yourself. ## Data protection - **Each organization's data is kept apart.** The database itself separates each organization's records, and the app runs as a role that can't switch that off. Work that has to cross organizations, such as key lookups and background jobs, runs under a separate, fenced role; any new use of it needs review, and its volume is monitored. - **Encrypted in transit.** Every connection to the site, the app and the API uses TLS. - **Stored credentials encrypted.** Tokens and keys for the accounts you connect are encrypted (AES-256-GCM) under versioned keys we can rotate without downtime, and they're deleted when you disconnect. API keys are stored only as hashes. - **Where your data lives.** Our servers, database and backups are in Germany, and backups are encrypted. The [subprocessor list](https://www.closefinch.com/subprocessors) shows what is processed elsewhere, and where. - **Agents ask first.** Emails, texts and other actions an agent takes outside the app pass the policy checks and, by default, wait for a person's approval. Content from outside (emails, web pages) can't make an agent send anything on its own. Website chat and voice agents answer live, within the limits you set. ## Access and identity - **Two-step sign-in.** With an authenticator app or a passkey, and backup codes. Owners need it to use owner settings, whichever way they sign in. - **Passkeys.** Sign in with a passkey instead of a password, or use one as the second step. - **Single sign-on and SCIM.** Sign in through your own identity provider (OIDC), and add and remove people with SCIM. Set up in Settings → SSO. - **Password checks.** Passwords found in known breaches are refused, and repeated failed sign-ins lock the account for a while. - **Sessions you can see.** A list of your signed-in sessions, and sign-out everywhere. - **Roles and access.** Owner, member and viewer roles, with access set by data source, agent, role and team, including which AI tools can see which data. ## Admin and audit - **Audit log.** Every change by a person or an agent is recorded, including each use of a connected account's credentials. The log is tamper-evident and checked every hour. - **Scoped API keys.** Each key gets only the scopes it needs, never more than the person who made it, and stops working when they leave. Sending messages to contacts needs an explicit send scope. - **Keys that expire.** Keys expire after 90 days by default (up to 365), with reminders before they do. A key that never expires is an owner's explicit, recorded choice. - **IP allowlists.** Limit a key to the addresses you list; a request from anywhere else is refused and recorded. - **Sign-in for scripts.** The command line signs in with a short code in your browser, so you don't need to paste a long-lived key. - **Monitoring and alerts.** Suspicious sign-ins, key misuse and spam raise alerts, and outbound sending can be stopped at once. ## Privacy - **Your data isn't sold or used to train models.** We don't sell it, and neither we nor our AI providers train their models on it. A feature you choose that makes something from your own material, such as a video replica of a teammate, uses only that material, with their written consent. - **Subprocessors and a DPA.** A published [subprocessor list](https://www.closefinch.com/subprocessors), and a [DPA](https://www.closefinch.com/dpa) with 30 days' notice before a new subprocessor. - **Export and deletion.** Owners can ask us for a full export of the organization's data. When you disconnect an account, its tokens are deleted, and an owner can also have everything that came from it deleted, search data included. Deleting the organization suspends it at once, lets an owner undo it for 30 days, then purges it. - **Staff access.** Our own staff's access is limited to what operating the service needs, and it's recorded. ## How we build and run it - **Secure development.** Dependencies are pinned and approved one by one, changes are checked by automated tests and a scan for leaked secrets, and each release passes an independent review before it ships. - **Incident response.** A written incident plan with drills on a schedule, and notice to you within 48 hours of a personal data breach that affects your data. Restore drills from backups are scheduled every quarter. ## Certifications None yet, and we don't show a badge we haven't earned. We plan SOC 2 when customers need it. [Contact us](mailto:security@closefinch.com) for our security questionnaire answers. ## Reporting a vulnerability Email security@closefinch.com with the details and how to reproduce them. We'll: - acknowledge within 2 business days; - keep you informed; - not pursue good-faith research that respects people's privacy and data, avoids disruption, and gives us reasonable time to fix before you disclose. Out of scope: denial-of-service, social engineering, physical attacks, and reports from automated scanners without a working proof. Machine-readable contact details are at [/.well-known/security.txt](https://www.closefinch.com/.well-known/security.txt). --- Source: https://www.closefinch.com/cookies # Cookie notice Last updated: November 8, 2026. - **This site.** The public pages (home, pricing, comparisons, docs and these legal pages) set no cookies and load nothing from other companies: no analytics scripts, no ads, no embedded widgets. If you arrive from another site or a campaign link, your browser keeps that site's name and the campaign's tags for 30 days (in local storage, not a cookie), and the Sign in and Start free links carry them to the app, so we can count where new teams come from. They're kept with your account only if you sign up. - **The app.** Once you sign in, the app uses only the cookies it needs to work: - a session cookie that keeps you signed in; - if you choose to trust a device at two-step sign-in, a cookie that remembers it for 30 days. They're first-party, `HttpOnly` and `Secure`, and they're never used for advertising or tracking. - **No banner.** We set no non-essential cookies, so we don't ask for cookie consent. If that ever changes, we'll ask first. How we handle personal data is in the [privacy policy](https://www.closefinch.com/privacy).