Humanloop logo
AI agent integrationHumanloop

Humanloop integration for multiplayer collaboration with AI agents using Claude Code or Codex

One governed connection your whole team and its AI agents can share, with approved actions and human review, so working in Humanloop never means pasting credentials into a prompt.

Use Humanloop from Claude Code

Bring Humanloop context into engineering work while Type keeps app access attached to the teammate and workspace.

Automate Humanloop with Codex

Let coding agents ask for the right app action, preserve conversation context, and keep humans in the approval loop.

Connect open agent workflows

Use Type as the collaboration layer around OpenClaw and other LLM workflows that need app access.

Developer Tools & IT

What the Humanloop integration exposes

Humanloop helps developers build and refine AI applications, offering user feedback loops, model training, and data annotation to iterate on language model performance

One connection, many teammates

Connect Humanloop once, then decide which teammates can use it for threads, automations, skills, and coding work.

Representative actions

  • 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. Upon execution, it returns details of the created project, including the project's id, name, description, created_at timestamp, and organization_id.

  • 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 linked to the project, will be permanently removed.

  • 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, description, creation timestamp, status, configuration details, and metrics/results. It is useful for monitoring experiments, analyzing results, tracking model configurations, and comparing experimental setups.

  • 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, reference_id, project information, datapoints_count, first_inputs, last_output, created_at, and updated_at. This enables users to monitor and analyze historical project interactions.

Connection

API and auth details

Humanloop exposes HTTP APIs and official Python and TypeScript SDKs for managing LLM application assets, prompts, logs, evaluations, datasets, experiments, feedback, deployments, and observability workflows.

FAQ

Questions people ask before connecting Humanloop

Can Claude Code use Humanloop?

Yes. Type lets an AI teammate use connected Humanloop actions from a governed workspace context, so Claude Code work can reference the app without copying credentials into a local prompt.

Can Codex work with Humanloop through Type?

Yes. Codex can collaborate through Type with app context, skills, and approved actions. The Humanloop catalog entry includes public integration details and example capabilities where available.

Is this the same as a Humanloop MCP server?

Type exposes connected app capabilities to AI teammates and coding agents through Type's integration layer. Teams use it when they want shared app access, human review, and teammate-level permissions around agent work.

More developer tools & it apps for AI teammates