# Humanloop — how to use (mcp.ai)

Connect your Humanloop account and use 4 tools for artificial intelligence straight from your AI agent. Connect with your own API key. Humanloop helps developers build and refine AI applications, offering user feedback loops, model training, and data annotation to iterate on language model performance.

## Option A — via MCP (recommended)
Remote MCP endpoint (HTTP, streamable): `https://api.mcp.ai/p_humanloop?ms=1787295660000`
Add it as a custom/remote MCP connector in your client (Claude, Cursor, VS Code…), then authenticate when prompted. Once connected, ask the agent to use the server's tools (e.g. `humanloop_create_project`).

## Option B — via direct REST API
Base URL: `https://api.mcp.ai/api/humanloop`
Auth: `Authorization: Bearer sk_live_…` — create a workspace API key at https://mcp.ai/settings/api-keys
Discover endpoints: `GET https://api.mcp.ai/api/humanloop/_endpoints`

### Endpoints
- `POST https://api.mcp.ai/api/humanloop/create/project` — This tool creates a new project in Humanloop. It is an independent action that generates a project by accepting a project's name (required), an optional description, and an optional organization_id. U
  - body: { name: string, description?: string, directory_id?: string }
- `POST https://api.mcp.ai/api/humanloop/delete/project` — This tool allows you to delete a specific project from your Humanloop organization. The deletion is permanent and cannot be undone. All associated data, including sessions, datapoints, and evaluations
  - body: { project_id: string }
- `POST https://api.mcp.ai/api/humanloop/list/experiments` — This tool retrieves an array of experiments associated with a specific project in Humanloop. It requires a project_id (starting with 'pr_') and returns details including experiment_id, name, descripti
  - body: { page?: integer, size?: integer, project_id: string }
- `POST https://api.mcp.ai/api/humanloop/list/sessions` — This tool retrieves a paginated list of sessions for a specific project in Humanloop. It requires a project_id (and optionally, page and size for pagination) and returns session details such as id, re
  - body: { page?: integer, size?: integer, project_id: string }

## Example prompts
- "What can I do in Humanloop?"
- "Show me a summary of my Humanloop account"

## More
- Page: https://mcp.ai/humanloop
- Agent spec (llms.txt): https://mcp.ai/humanloop/llms.txt
- Postman collection: https://mcp.ai/humanloop/postman.json