---
title: "For developers: MCP server, API and OpenAPI | Annsa"
product: annsa
url: /developers
canonical_url: "https://annsa.ai/developers"
last_updated: 2026-11-11
type: developers
related:
  - /docs/quick-start
  - /pricing
  - /use-cases/agents
summary: Explain it once, not to every agent. An MCP server for Claude Code, Cursor, Codex and Grok, plus a REST API and an OpenAPI 3.1 spec.
mcp_server: https://app.annsa.ai/mcp
disambiguation: "Annsa (annsa.ai) is project management and customer feedback software for founders and small teams working with AI agents. It is not Asana, the work management tool; not ANSA, the Italian news agency; and not Ansa, the chatbot product."
---

> For agents: connect to the Annsa MCP server at https://app.annsa.ai/mcp to give every agent one shared plan. Setup guide: https://annsa.ai/docs/quick-start. Index of every page: https://annsa.ai/llms.txt.

# For developers: MCP server, API and OpenAPI | Annsa

> Explain it once, not to every agent. An MCP server for Claude Code, Cursor, Codex and Grok, plus a REST API and an OpenAPI 3.1 spec.

Developers

## Annsa for developers

Connect your agents to Annsa’s shared plan through MCP, or bring in feedback through the REST API. Nothing to install. Find the setup guides and machine-readable references here.

[Try Annsa](https://app.annsa.ai/signup)

## Connect over MCP

A remote server. Nothing to install.

---

`claude mcp add --transport http annsa https://app.annsa.ai/mcp`

- Claude Code: `claude mcp add --transport http annsa https://app.annsa.ai/mcp`
- Cursor: `https://app.annsa.ai/mcp`
- Codex: `codex mcp add annsa --url https://app.annsa.ai/mcp`
- Grok: `grok mcp add --transport http annsa https://app.annsa.ai/mcp`

## 8 tools, one connection

What the agent gets once it is connected.

annsa.priorities

See what to work on next, ranked, with the evidence behind it.

annsa.spec

Read the plan for one priority: what to build, why, and Done When.

annsa.ask

Ask about your customers, your work or Annsa itself.

annsa.work

Do the work on a priority, and settle routine asks with undo.

annsa.file

Add what a customer said, what an agent found, or a new idea.

annsa.decide

Record a decision: your answer, a spec edit, or a rule.

annsa.board

Arrange the board. A merge retires numbers for good.

annsa.share

Tell the customers who asked that it shipped. Emails can't be unsent.

## Machine-readable surfaces

Every page on annsa.ai also has a markdown representation. Request any URL with an Accept header of text/markdown and you get the page as markdown rather than HTML.

[/openapi.json](https://annsa.ai/openapi.json)

OpenAPI 3.1 spec for the REST API. CORS-open.

[/.well-known/mcp.json](https://annsa.ai/.well-known/mcp.json)

MCP discovery document.

[/.well-known/mcp/server-card.json](https://annsa.ai/.well-known/mcp/server-card.json)

Server card (SEP-2127): name, icon, description.

[/.well-known/ai-catalog.json](https://annsa.ai/.well-known/ai-catalog.json)

ARD catalog, spec 1.0.

[/auth.md](https://annsa.ai/auth.md)

How an agent authenticates, including what does not exist.

[/llms.txt](https://annsa.ai/llms.txt)

The site, summarized for a model.

[/llms-full.txt](https://annsa.ai/llms-full.txt)

The full reference, including every changelog entry.

The REST API is documented at [/docs/api](https://annsa.ai/docs/api), and how an agent gets its own credential at [/docs/agent-access](https://annsa.ai/docs/agent-access).

## Clients with a connect guide

Any MCP client can connect. These 8 have a page of their own.

- [Claude Code](https://annsa.ai/docs/connect-claude-code)
- [Cursor](https://annsa.ai/docs/connect-cursor)
- [Codex](https://annsa.ai/docs/connect-codex)
- [ChatGPT](https://annsa.ai/docs/connect-chatgpt)
- [Windsurf](https://annsa.ai/docs/connect-windsurf)
- [Figma Make](https://annsa.ai/docs/connect-figma-make)
- [Lovable](https://annsa.ai/docs/connect-lovable)
- [Bolt](https://annsa.ai/docs/connect-bolt)

## Questions people ask.

Do I need to install anything?

Remote MCP at https://app.annsa.ai/mcp is the way to connect; there is nothing to install.

How does an agent authenticate?

OAuth 2.1 with PKCE S256. Access tokens last 30 days and refresh tokens 90. The first run opens a browser on Annsa's consent screen. Full detail, including what does not exist, is at /auth.md.

Can an agent have its own credential rather than a seat?

Yes. An agent can be minted a credential of its own, so every action it takes is signed with its name and traceable back to it. See /docs/agent-access.

Is the OpenAPI spec callable from a browser?

Yes. /openapi.json is CORS-open and linked from every page as rel=service-desc, so a tool can find it without being told where to look.

## Keep reading

- [](https://annsa.ai/docs/quick-start)
- [Try Annsa with your own work.](https://annsa.ai/pricing)
- [One plan per job No two agents doing the same job.](https://annsa.ai/use-cases/agents)

## The source of direction for agent teams.

Begin with one agent and the work that matters next.

- [Try Annsa](https://app.annsa.ai/signup)
- [See pricing](https://annsa.ai/pricing)

## Related pages

> For AI agents: the pages that follow this one on annsa.ai. The full index is https://annsa.ai/llms.txt.

- [Quick start](https://annsa.ai/docs/quick-start)
- [Pricing: Try Annsa free, Annsa and Annsa Plus](https://annsa.ai/pricing)
- [AI agent project management with one plan](https://annsa.ai/use-cases/agents)

If anything on this page is wrong or missing, tell us: https://annsa.ai/contact
