Metadata-Version: 2.4
Name: sphinx-agentready
Version: 1.0.0
Summary: Sphinx extension that submits your docs to AgentReady after every build, making them instantly queryable by AI agents via MCP.
Project-URL: Homepage, https://www.agentready.it.com
Project-URL: Repository, https://github.com/AshutoshRaj97/agentready-mcp/tree/main/sphinx-plugin
Project-URL: Bug Tracker, https://github.com/AshutoshRaj97/agentready-mcp/issues
License: MIT
Keywords: agents,ai,documentation,extension,llms,mcp,sphinx
Classifier: Development Status :: 4 - Beta
Classifier: Framework :: Sphinx :: Extension
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Documentation
Classifier: Topic :: Software Development :: Documentation
Requires-Python: >=3.8
Requires-Dist: sphinx>=5.0
Description-Content-Type: text/markdown

# sphinx-agentready

Sphinx extension that submits your docs site to [AgentReady](https://www.agentready.it.com) after every build, making it instantly queryable by AI agents (Claude, Cursor, Windsurf, and any MCP client) with cited, multi-page answers.

## Installation

```bash
pip install sphinx-agentready
```

## Usage

Add the extension to your `conf.py`:

```python
extensions = [
    "sphinx_agentready.extension",
]

# Domain is inferred from html_baseurl automatically, or set explicitly:
agentready_domain = "docs.yoursite.com"
```

After each `sphinx-build` or `make html`:

```
[AgentReady] ✓ Indexed docs.yoursite.com. AI agents can now query it.
[AgentReady]   MCP:  https://www.agentready.it.com/api/mcp
```

## How it works

After your site builds, the extension posts your domain to AgentReady's indexing webhook. AgentReady crawls your docs and makes them queryable via:

- **MCP server** — `https://www.agentready.it.com/api/mcp` (works in Claude, Cursor, Windsurf, and any MCP client)
- **REST endpoint** — `POST https://www.agentready.it.com/api/sites/{id}/ask`
- **llms.txt** — `https://www.agentready.it.com/api/sites/{id}/llms.txt`

## Connecting to your AI tools

**Claude Code / Claude Desktop**

```bash
claude mcp add agentready npx @agentreadyweb/mcp
```

**VS Code (GitHub Copilot)**

```json
// .vscode/mcp.json
{
  "servers": {
    "agentready": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@agentreadyweb/mcp"]
    }
  }
}
```

Then ask your AI: _"Ask agentready: how do I configure X in [your docs]?"_

## Configuration

| Option | Default | Description |
|--------|---------|-------------|
| `agentready_domain` | _(inferred from `html_baseurl`)_ | Domain of your docs site (e.g. `docs.example.com`) |

## License

MIT © [AgentReady](https://www.agentready.it.com)
