Anvil CRM

npx add-mcp 'https://app.anvilcrm.com/api/mcp'

Turn CRM context into useful action while keeping control.

MCP gives AI assistants a structured way to interact with Anvil. Your team can use a compatible assistant to read, create, update, and analyze CRM records while Anvil's existing workspace permissions keep access scoped to the right people and data.

Works with the assistants your team already uses

What an assistant can do with Anvil

• Work from real CRM context

  Let assistants use the accounts, contacts, leads, deals, products, and activities already in your workspace.

• Build useful sales workflows

  Ask for reports, identify next actions, research account context, and create workflows that fit the way your team sells.

• Keep access permission-aware

  MCP actions respect the workspace access and permissions configured in Anvil.

A CRM that exposes defined tools, not an unrestricted database connection.

The Model Context Protocol gives an AI client a standard way to discover and call tools. Anvil's remote MCP server describes CRM objects, readable fields, filters, and allowed actions. The assistant receives structured results instead of browser access or database credentials.

Every connection is bound to one workspace during OAuth consent. Anvil then intersects the token scopes with the stored grant, workspace switches, plan entitlements, role permissions, record visibility, and pipeline access on every call. A scope can narrow access; it cannot grant authority the member does not already have.

An AI CRM that works with the assistant you already use.

Most AI CRM software builds its own assistant into the product and charges for it through credits or add-ons. Anvil does not. It gives ChatGPT, Claude, and other MCP clients a controlled way into your CRM, so your team keeps the assistant it knows and pays for it once.

Use these six questions to compare any AI CRM. The answers show how Anvil handles each one.

Does the AI see only what the user may see?	Yes. Every call runs as the member who connected the assistant, with that member's live permissions.
Can you choose the AI model?	Yes. Use ChatGPT, Claude, Cursor, Codex, or another MCP client. Anvil does not lock you into one model.
Can you limit what the AI may change?	Yes. Grant read, write, or delete scopes separately, and keep workspace writes switched off if you want read-only access.
Is every AI action recorded?	Yes. Each tool call records the member, client, tool, outcome, and touched record IDs.
Can you remove access at once?	Yes. A member or an admin can revoke a connection, and the next call fails.
How is the AI priced?	MCP access is part of the Pro price per user. There are no AI credits. Your team uses the AI subscription it already has.

Connect a compatible client in four controlled steps.

1. An admin enables agents

   On Pro, a workspace admin turns on agent access under Settings → AI agents. Write access remains a separate switch and can stay off.

2. Add the remote endpoint

   Configure the client with https://app.anvilcrm.com/api/mcp. Anvil publishes OAuth authorization and protected-resource metadata for discovery.

3. Choose workspace and scopes

   Sign in, select exactly one workspace, and approve read, write, correction, or delete scopes. The client never chooses a tenant from unverified input.

4. Inspect before acting

   Call whoami first to see the workspace, effective scopes, CRM access, and permission matrix. Use discovery tools before sending record fields.

Remote MCP endpoint

https://app.anvilcrm.com/api/mcp

Grant only the class of action the workflow needs.

anvil:read	Object discovery, records, activities, timelines, pipeline summaries, and reports	Results still follow record, field, pipeline, and module permissions
anvil:write	Create and update CRM records, activities, relationships, lifecycle actions, audited closure corrections, and commercials	Disabled when the workspace write switch is off; writes use revision or idempotency controls
anvil:delete	Delete a record or activity occurrence	Requires an explicit confirmation flag; CRM record deletion is soft deletion

Move from discovery to a fenced CRM action.

• Discover

  Use list_objects and describe_object to learn available object types, fields, list columns, filters, and write contracts instead of guessing payloads.

• Read and diagnose

  Search or list records, inspect activities and timelines, get pipeline summaries, and find work that needs attention within the caller's visibility.

• Act with concurrency controls

  Create records and tasks with idempotency keys. Update existing records with their expected revision so a stale assistant cannot overwrite a newer human change.

• Handle lifecycle explicitly

  Preview stage, close, reopen, conversion, ownership, and commercial actions before applying them. Dedicated tools preserve required values and lifecycle history.

Give the assistant a goal and a review boundary.

• “Show the open deals I can see that have no completed activity in the last 14 days. Do not change anything.”

• “Summarize the visible timeline for Meridian Systems and list unresolved next actions with their owners.”

• “Preview moving this deal to Proposal and tell me which values or permissions are missing. Do not apply the change.”

• “Draft five follow-up tasks from this pipeline review. Show the proposed titles, owners, and due dates before creating them.”

Safeguards and limitations to understand before connecting.

• Live permissions

  Every call re-checks membership, roles, record access, pipeline visibility, plan entitlements, scopes, and workspace agent switches.

• Explicit high-impact confirmation

  Deletion and workspace-structure tools require the assistant to restate the change, obtain confirmation in the same turn, and send confirm: true.

• Auditable agent activity

  Tool calls record the grant, member, client, tool, outcome, duration, and touched record IDs, but not prompts, inputs, results, or field values.

• Immediate revocation

  A member can revoke their grant and an admin can revoke any workspace grant. Disabling workspace access or removing membership also blocks calls.

• Bounded requests

  Lists are paginated, request and response sizes are capped, rate limits apply, and writes use revision fences or idempotency keys.

• Human review still matters

  The server enforces authorization and data contracts; it cannot guarantee that an assistant's interpretation, summary, or proposed business decision is correct.

Sales teams already running on Anvil

Questions and answers

What is an AI CRM?

An AI CRM lets an AI assistant work with your customer data: summarize accounts, find deals that need attention, draft follow-up tasks, and update records. In Anvil, the assistant works through MCP and stays inside each member's permissions.

Does Anvil include its own AI model?

No. Anvil works with the AI assistant your team already uses, such as ChatGPT or Claude, through MCP on Pro. You choose the model, and Anvil enforces what it can see and change.

What is an MCP CRM?

An MCP CRM connects to AI assistants through the Model Context Protocol, giving them structured access to CRM tools and data within defined permissions.

Which AI assistants work with Anvil MCP?

Anvil MCP uses the remote streamable HTTP transport and works with compatible clients including ChatGPT, Claude.ai, Claude Code, Cursor, and Codex. Client support changes independently, so confirm that your chosen client supports remote MCP servers and OAuth.

What can an AI assistant do with Anvil CRM?

Depending on its OAuth scopes and the member's live permissions, an assistant can inspect CRM structure, search and read records, review pipelines and timelines, create or update records and activities, run reports, and perform explicitly confirmed structure or deletion actions.