Relm

ContactsCompaniesDeals ActivitiesPipelinesAutomationsWebhooks

This hits Relm for real, in free test mode. Edit the values and go.

No boilerplate, no schema migration, no sales-ops onboarding. Mint a key, point your agent at it, go.

Sign up and mint arelm_live_key. Test-mode keys are free and their data is invisible to billing - build against them freely (test records auto-delete 7 days after creation).

Drop the MCP URL into Claude or hit the REST API from any language. Your agent reads the live schema first, so it is never confused about what fields or values exist.

The agent creates contacts, moves deals across pipelines, logs activities and fires automations - the same primitives a human CRM has, exposed as clean function calls.

Everything a real CRM needs. Nothing bolted on.

A complete object model, addressable as an API. Every endpoint has a matching MCP tool.

People with names, emails, phones, LinkedIn, type and custom fields. Email-optional - phone or LinkedIn-only leads are first-class.

Accounts. Link contacts and deals to the organizations they belong to; filter and search by domain.

Opportunities across multiple named pipelines, each with its own ordered stages. Moves emit events.

One timeline for notes, calls, emails and meetings - with backdating, so imported history keeps its real dates.

Event-triggered rules and multi-step drip sequences. Move a stage, log an activity, set metadata, send an email.

Signed HMAC deliveries with retry and a dead-letter queue. Fan CRM events out to the rest of your stack.

Bring your own Resend key (encrypted per workspace); automations send email via thesend_emailaction. Test-mode sends are simulated, never delivered.

A live, self-describing schema. Unknown values return valid options and a did-you-mean, so agents self-correct instead of failing blind.

A mode flag on every object. Test data is free, unmetered and invisible to your dashboard and billing.

Designed to be discovered and driven by AI

Relm is not a UI with an API stapled on. It is machine-first from the DNS record up.

A full Model Context Protocol server - every CRM operation is an MCP tool with a typed schema. Batch many tool calls in one round-trip. Point any MCP client atapi.relmcrm.com/mcp.

Every error is RFC-9457 problem+JSON withvalid_options. An agent that sends a bad enum gets back the list of good ones - it fixes itself without a human.

/llms.txt,/.well-known/mcp.json, an A2A agent card and an AI-crawler-friendlyrobots.txt. Agents can find and understand Relm on their own.

Prefixed IDs (con_,deal_), cursor pagination, anIdempotency-Keyheader and versioned optimistic concurrency. The boring guarantees agents rely on.

Why not just a Postgres table or HubSpot's API?

Because your agent needs CRM behavior, not just storage - and it needs an interface built for a machine, not a sales rep.

Start free. Pay when your agents get busy.

Priced per API request, not per seat - because your main user is a machine. Overage is metered, and you can set a hard spend cap.

Overage is billed at $0.0001 / request. Set a spend cap to $0 and Relm hard-stops instead of billing you.

Relm is an API-first CRM built for LLMs and AI agents. It exposes contacts, companies, deals and activities over a REST API and a native MCP server, so an AI agent with an API key can read and write your CRM directly. The web dashboard is a thin human view for the times you want to look.

A CRM whose primary user is a language model, not a sales rep. Instead of forms and pipelines you click through, the agent drives everything through function calls: it creates records, moves deals, logs activities and triggers automations. Relm is designed so an LLM is never confused about your schema.

Both, over the same data and the same API key. MCP clients like Claude connect tohttps://api.relmcrm.com/mcpand get every operation as a typed tool. Everything else uses the REST API athttps://api.relmcrm.com/v1.

Yes - 1,000 API requests per month, no credit card. Test mode is always free and unmetered, so you can build and demo without touching your quota.

How is this different from HubSpot, Attio or Pipedrive?

Those are UI-first CRMs for human reps, with an API added later. Relm inverts it: the API and MCP server are the product; the dashboard is the thin part. If your "sales team" is a set of agents, a UI-first CRM fights you and an API-first one gets out of the way.

Yes. Connect the MCP server to Claude (or any MCP client) and run the whole CRM in natural language - "import these 200 leads", "move everyone stuck in demo for 30 days to nurture". For custom apps, the REST API is a few lines.

Yes. Pipelines are first-class - each workspace has one or more, each with its own ordered stages. The same stage key (likewon) can live independently in several pipelines, and every deal carries its pipeline, stage label and semanticstage_type(open/won/lost) so an agent never has to guess.

Two doors. End-user clients (Claude directory, ChatGPT) connect throughOAuth 2.1 sign-in- PKCE, no key handling. Scripts and custom agents use a bearer key:Authorization: Bearer relm_live_.... Keys are workspace-scoped, shown once, SHA-256 hashed at rest, and come inliveandtestvariants (OAuth covers live mode; test stays key-only).

Sign up, mint a key, and let it run. Free to start, no credit card.

Free MCP that drives an audit of your marketing. Your AI connects, adsOS digs through your ads, email and site, and hands back a growth plan you can run today.

Talk to your CRM's AI agent from Claude, Cursor, and other MCP clients — read records, send email, create reminders, draft outreach, all gated by approvals.

Customer Intelligence & Segmentation

Find Which next customer to call and how much will be worth it

Operate and customize a tenant-isolated AI CRM for customers, sales, jobs, marketing, reviews, workflows, documents, and workspace pages.

Find and enrich B2B leads, run multi-channel outreach and manage the sales pipeline — 27 annotated tools.

Run your entire sales outreach by chatting with AI

Find buyers and book more meetings without your AI Assistant

Interact with data in Attio, the AI-native CRM, enabling AI assistants to access and manage your customer relationship information.

A production-ready MCP server for Customer Relationship Management (CRM) functionality, built with TypeScript and SQLite.

Interact with Microsoft Dynamics 365 for customer relationship management.