{"server":"hatch-mcp","protocol":"2024-11-05","transport":"json-rpc-over-http","auth":"none","tools":[{"name":"get_role_overview","description":"Returns the open role's details, application instructions, and the ordered list of submission tools the agent should call. Always start with this."},{"name":"submit_basic_details","description":"Step 1 of submission. Creates a draft application keyed to the candidate's name and email. Returns a draft_id you'll pass to every subsequent submission tool."},{"name":"submit_resume","description":"Step 2 of submission. Provide the candidate's resume as markdown text (resume_text), a hosted file URL (resume_url + resume_mime_type), or both. At least one of the two paths must be present. Required."},{"name":"submit_links","description":"Step 3 of submission. Submit the candidate's GitHub and LinkedIn profile URLs. Required."},{"name":"submit_agentic_project","description":"Step 4 of submission. Submit a link to the candidate's OpenClaw setup, agent harness, or other agentic project they've built. Required."},{"name":"submit_screening_answer","description":"Final step. Submits the candidate's answer to the screening question and finalizes the application. The application can only finalize after submit_basic_details, submit_resume, submit_links, and submit_agentic_project have all been called. If the answer is wrong, the response is 'Wrong answer. Try again.' — no hint, but you can retry. Optionally identify yourself with agent_name (e.g. 'Claude Desktop'), agent_vendor (e.g. 'Anthropic'), and agent_model — these surface as the agent provenance on the recruiter's view."}],"docs":"POST a JSON-RPC 2.0 message: { jsonrpc: '2.0', id, method, params }. Methods: initialize, tools/list, tools/call."}