Metadata-Version: 2.5
Name: agimus-mcp
Version: 1.2.0
Summary: MCP server for the Agimus data platform — connects AI coding tools to your ontology
Project-URL: Homepage, https://agimus.ai
Project-URL: Repository, https://github.com/agimus-ai/agimus-platform
Project-URL: Issues, https://github.com/agimus-ai/agimus-platform/issues
Author-email: Agimus <support@agimus.ai>
License: MIT
Keywords: agimus,ai,data-platform,llm,mcp,ontology
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Software Development :: Libraries
Requires-Python: >=3.10
Requires-Dist: agimus<3,>=2
Requires-Dist: mcp<3,>=2.0
Requires-Dist: pydantic>=2.7
Description-Content-Type: text/markdown

# Agimus MCP

A read-only MCP server for discovering an Agimus ontology and querying its objects. It provides 13 tools for schema discovery, paged reads, counts, distinct values and aggregates, plus live entity-schema and bundled Python SDK documentation resources.

The server runs over **stdio** and provides ontology discovery and read tools. Use the Agimus SDK for object writes, SQL queries and Arrow results.

Sign in to the Agimus app and open **Resources → SDK & API Docs → MCP** for all tool arguments, results and worked examples. The same page contains the SDK guide, Python reference and HTTP API reference.

## Connect

Create a read-only API key in your Agimus tenant. With `uvx` available on your PATH, use this generic MCP configuration in your client:

```json
{
  "mcpServers": {
    "agimus": {
      "command": "uvx",
      "args": ["agimus-mcp"],
      "env": {
        "AGIMUS_API_KEY": "agm_your_key",
        "AGIMUS_BASE_URL": "https://api.agimus.ai"
      }
    }
  }
}
```

Adapt the enclosing configuration to your MCP client's format. For demo, use a demo API key and `https://demo-api.agimus.ai`. Keep credentials out of source control. Call `agimus_whoami` first to confirm the tenant.

Alternatively, `pip install agimus-mcp` and configure the client to run `agimus-mcp` with no arguments. Requires Python 3.10 or later; the Agimus SDK dependency is installed automatically.

## Options

| CLI flag | Environment variable | Default |
| --- | --- | --- |
| `--api-key` | `AGIMUS_API_KEY` | Required |
| `--base-url` | `AGIMUS_BASE_URL` | `https://api.agimus.ai` |
| `--timeout` | `AGIMUS_TIMEOUT` | `30` seconds |
| `--log-level` | `AGIMUS_LOG_LEVEL` | `WARNING` |

Logs use stderr; stdout carries the MCP protocol. Read tools offer Markdown or JSON text output. Pages contain at most 500 objects; check continuation cursors and aggregate/distinct truncation before treating results as complete.

Read the `agimus://docs/sdk-reference` resource for the bundled SDK guide and reference. Live schemas are available at `agimus://schema/entities/{entity}`.

## License

MIT
