> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.meetstream.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.meetstream.ai/_mcp/server.

# Docs for Agents & LLMs

> Use MeetStream docs from AI agents: llms.txt index, .md versions of every page, the docs-search and action MCP servers, agent-friendly CLI output, and the published OpenAPI spec.

These docs are built to be read by agents as much as by humans. If you're integrating MeetStream with Claude, Cursor, or any LLM-driven workflow, start here.

## The machine-readable surface

| Resource                  | URL                                                                        | What it's for                                                                 |
| ------------------------- | -------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
| llms.txt index            | [docs.meetstream.ai/llms.txt](https://docs.meetstream.ai/llms.txt)         | The page index agents should fetch first                                      |
| Markdown twin of any page | append `.md` to any docs URL                                               | Clean Markdown, no HTML chrome — e.g. `docs.meetstream.ai/guides/help/faq.md` |
| OpenAPI spec              | [docs.meetstream.ai/openapi.json](https://docs.meetstream.ai/openapi.json) | The full API schema for codegen and validation                                |

## Two MCP servers (use both)

* **Docs-search MCP** — `https://docs.meetstream.ai/_mcp/server` — searches these docs. Read-only, no key needed.
* **[Action MCP](/build-with-ai/meetstream-mcp-server)** — `https://mcp.meetstream.ai/mcp` (hosted, auth via your key) or `npx @meetstream/mcp` (local) — actually **does things**: creates bots, pulls transcripts and status, sends messages into live meetings. This is the one that lets an agent debug a real bot session.

Claude Code users can install both in one step with the plugin:

```bash
claude plugin marketplace add meetstream-ai/meetstream
```

## Agent-friendly CLI

Every [`@meetstream/cli`](/build-with-ai/meetstream-cli) command supports `--json`, so agents can drive the full API from a shell without parsing prose:

```bash
meetstream bot create "https://meet.google.com/abc-defg-hij" -t deepgram --json
meetstream bot status <bot_id> --json
meetstream transcript <bot_id> --wait --json
```

## Ground rules worth telling your agent

* Auth is `Authorization: Token <key>` — **`Token`, not `Bearer`**. The word matters.
* `create_bot` requires `meeting_link` AND `bot_name`.
* Webhook envelopes carry both `event` and `bot_event`; branch on `bot_event`/`bot_status` for terminal reasons — details in [Debugging Bots](/guides/help/debugging-bots).
* An `Idempotency-Key` replay returns **507 with the original bot** — treat as success, never retry-loop on it.
* Media links are presigned and expire — re-fetch the endpoint, don't cache URLs.