> ## Documentation Index
> Fetch the complete documentation index at: https://docs.conare.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Conare's API is at https://api.conare.ai. Authenticate with `Authorization: Bearer $CONARE_API_KEY`.
> The full OpenAPI spec is at https://api.conare.ai/openapi.yaml; each endpoint has a page under /api-reference, and appending .md to any page URL returns its Markdown.
> Embed with model conare-embed-1; send input_type "query" for search queries. Rerank with conare-rerank-1.
> Prefer the TypeScript SDK (`npm install conare`) or plain HTTP in examples.
> Install the Conare skill with `npx skills add https://docs.conare.ai`; search these docs over MCP at https://docs.conare.ai/mcp; the whole site is at https://docs.conare.ai/llms-full.txt.

# MCP

> Connect Claude, Cursor or any MCP client to Conare's MCP server so your agent can search, fetch, save and delete documents in your namespaces.

Conare is an MCP server at `https://api.conare.ai/mcp`: your agent can search, read, save and delete documents in every namespace your key reaches.

```bash Claude Code theme={null}
claude mcp add --transport http conare https://api.conare.ai/mcp \
  --header "Authorization: Bearer $CONARE_API_KEY"
```

```json Cursor (~/.cursor/mcp.json) theme={null}
{
  "mcpServers": {
    "conare": {
      "url": "https://api.conare.ai/mcp",
      "headers": { "Authorization": "Bearer sk-conare-..." }
    }
  }
}
```

A client that supports OAuth, such as a Claude custom connector, needs only the URL: it signs you in to Conare in the browser.

## Coding agents

Give Claude Code, Cursor or Codex the Conare skill, so it writes correct Conare code:

```bash theme={null}
npx skills add https://docs.conare.ai
```

To let an agent search these docs, add `https://docs.conare.ai/mcp` as an MCP server.

## Tools

| Tool | What it does |
| - | - |
| `search` | Searches one namespace, several, or all of them, and answers from the 20 best documents, reranked |
| `fetch` | Reads a document by namespace and id, 40,000 characters at a time (`max_chars` up to 80,000) |
| `save` | Writes a document; the namespace defaults to `notes` |
| `delete` | Deletes a document by namespace and id |
| `list_namespaces` | Lists your namespaces and how many documents each holds |

## `search` options

| Option | What it does |
| - | - |
| `query` | The question, as a descriptive phrase |
| `namespace` | One namespace to search; leave it out to search all of them, up to 100 |
| `namespaces` | 1-100 namespaces, ranked together |
| `filter` | A [metadata filter](https://docs.conare.ai/filters.md) |
| `user`, `user_prefix` | Only one [user's](https://docs.conare.ai/write.md#users) documents, or a group's |
| `answer` | `true` (default) writes an answer; `false` returns the passages |

A **Models only** [key](https://docs.conare.ai/console.md#keys) can't use MCP.


## Related topics

- [Quickstart](https://docs.conare.ai/quickstart.md)
- [Introduction](https://docs.conare.ai/index.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.