# humwork

Humwork is a real-time consultation marketplace that connects AI agents to verified human domain experts. Open a session with consult_expert, exchange messages until the problem is resolved, then clo…

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y smithery mcp add humwork/humwork

# Browse available tools
npx -y smithery tool list humwork/humwork

# Get full schema for a tool
npx -y smithery tool get humwork/humwork consult_expert

# Call a tool
npx -y smithery tool call humwork/humwork consult_expert '{}'
```

## Direct MCP Connection

Endpoint: `https://humwork--humwork.run.tools`

## Tools (6)

- `consult_expert` — Consult a verified domain expert in real-time. Creates an interactive chat session where the expert diagnoses your prob…
- `send_chat_message` — Send a message in an active expert chat session. Use this to respond to expert questions, share additional context, or …
- `get_chat_messages` — Retrieve messages from an expert chat session. Use since_message_id to get only new messages you haven't seen yet.
- `extend_session` — Extend an active expert chat session by pre-paying for another 10-minute block. Call this when you receive a grace_peri…
- `close_chat` — Close an active expert chat session. This ends billing, frees the expert for other work, and processes final payment.
- `rate_chat` — Rate an expert's helpfulness after a chat session has closed. Use this to:

```bash
# Get full input/output schema for a tool
npx -y smithery tool get humwork/humwork <tool-name>
```

## Prompts (3)

- `assess_progress` (problem, attempts_so_far) — Evaluate whether your current approach is making progress or whether expert consultation would be more efficient. Use t…
- `review_my_approach` (decision, options, context) — Get your approach, strategy, or important decision reviewed by a domain expert before committing to it. Use this when t…
- `prepare_consultation` (problem, context, errors, attempted_solutions) — Helps you structure a clear, well-organized expert consultation request. Use this before calling consult_expert to ensu…