Expedify over MCP

Expedify speaks MCP (the Model Context Protocol), so an AI app or agent can drive the whole product. Search your CRM, author and run workflows, query analytics, ask your knowledge base, as tools the model calls directly.

There are two ways in:

• Sign in from ChatGPT, Claude or Gemini. Each person connects their own account. The app works as them, sees only the records they can see, and does only what they allowed on the consent screen.
• A scoped API key for Claude Code, Cursor, n8n or your own agents. The key works for the whole organisation, within the scopes it was given.

MCP is a co-equal surface with the REST API, not an add-on. Both are projections of the same core:

                    ┌──────────────┐
   REST  /v1/*  ───▶│              │
                    │  ONE core:   │   entity registry · scopes · services
   MCP  /v1/mcp ───▶│  operations  │
                    │  .execute()  │
                    └──────────────┘

Every guarantee you get on REST — the agx_ key, least-privilege scopes, default-deny, tenant isolation, rate limits — applies identically on MCP, because a tool call runs through the exact same gate as the REST endpoint.

Why MCP (and not just REST)

• Agent-native. Tools are curated intents (search_contacts, run_workflow, ask_knowledge_base) that read like instructions — not a 1:1 mirror of every REST route. A model picks the right tool without you writing glue code.
• One permission model. Sign-in and API keys use the same scopes. They decide which tools may run (the catalog itself is public).
• No SDK needed. Chat apps connect with an address and a sign-in, and developer tools with a couple of lines of config (see Connect).

The mental model

	REST	MCP
Transport	HTTPS request/response	JSON-RPC (tools/list / tools/call)
Endpoint	https://api.expedify.ai/v1/...	https://api.expedify.ai/v1/mcp, or /v1/mcp/<org-slug> for one organisation
Auth	Authorization: Bearer agx_…	Sign-in (OAuth) from chat apps, or Authorization: Bearer agx_… (same key)
Unit	an endpoint + method	a tool
Shape	fine-grained CRUD	curated agent intents
Discovery	this OpenAPI reference	tools/list (full catalog; scopes enforced at call)

Start here

1. Connect: add Expedify to ChatGPT, Claude or Gemini, or point Claude Code, Inspector or n8n at the server.
2. Auth & scopes: how sign-in permissions and key scopes gate which tools can run (discovery is public).
3. Tool catalog — every tool, grouped by module, with schemas + examples (generated from the server, so the count there is always current).
4. Workflow tools — expose one of your own workflows as its own run_… tool.
5. Recipes — worked end-to-end flows (author a workflow, ask a KB, NL→SQL→chart).

Discovery is public; execution is not

tools/list needs no auth (it returns the static catalog, with no org data). Execution (tools/call) requires a sign-in or a key, and every call is scope-, tenant- and ACL-checked.