{
  "markdown": "<p align=\"center\">\n  <a href=\"https://humanhours.dev\">\n    <img src=\"https://humanhours.dev/icon.svg\" alt=\"HumanHours\" width=\"96\" height=\"96\" />\n  </a>\n</p>\n\n<h1 align=\"center\">HumanHours MCP Server</h1>\n\n<p align=\"center\">\n  <em>Every agent, measured in human hours.</em><br/>\n  <a href=\"https://humanhours.dev\">humanhours.dev</a> ·\n  <a href=\"https://humanhours.dev/docs/mcp\">MCP docs</a>\n</p>\n\nTrack agent ROI and enrich companies straight from any MCP client. HumanHours is a\nremote, hosted MCP server: connect once with your HumanHours account and your agents\ncan log task events, read what they saved, and pull real labour-cost data for any company.\n\nNo installation, no API keys to copy around, no local process to run.\n\n## Connect\n\n| | |\n|---|---|\n| **Endpoint** | `https://mcp.humanhours.dev/mcp` |\n| **Transport** | Streamable HTTP |\n| **Auth** | OAuth 2.1 (PKCE) — browser login with your HumanHours account, approve workspace access |\n\nAdd it to any client that supports remote MCP servers (Claude Desktop, claude.ai,\nCursor, and others). On first use you complete a browser login and approve the scopes\nbelow; the client stores a token, so there are no keys to manage by hand.\n\n### Claude Desktop / claude.ai\n\nAdd a remote connector pointing at `https://mcp.humanhours.dev/mcp` and complete the\nlogin when prompted.\n\n### Cursor / other clients\n\n```json\n{\n  \"mcpServers\": {\n    \"humanhours\": {\n      \"url\": \"https://mcp.humanhours.dev/mcp\"\n    }\n  }\n}\n```\n\n## Tools\n\n**Logging**\n- `track_event` — Record one agent task event and compute the hours and money it saved.\n\n**Reading**\n- `get_summary` — Aggregated ROI for a period, with a daily trend.\n- `time_saved_report` — Hours saved, grouped by agent, task type, model, client, or time.\n- `cost_saved_report` — Money saved, grouped the same ways.\n- `leaderboard` — Top agents by a chosen metric.\n- `get_usage` — Event usage against the plan's included quota.\n- `list_agents` — Agents tracked, with 30-day hours and money saved.\n- `list_task_types` — Built-in plus custom task types.\n\n**Enrichment**\n- `enrich_company` — Turn a domain into roles, per-hour wages, and an annual labour-cost case.\n- `get_company` — Read one enriched company from the library.\n- `list_companies` — List the workspace's company library, optionally by tag.\n- `bulk_enrich` — Queue a background job for many domains; returns a job id.\n- `get_job` — Check the status of a bulk enrichment job.\n\n## Scopes\n\n| Scope | Grants |\n|---|---|\n| `track:write` | Log task events |\n| `read:reports` | Read ROI summaries, reports, leaderboards, usage |\n| `enrich:write` | Enrich companies and read the company library |\n\n## Docs\n\nFull connection guide, scopes, and tool reference: https://humanhours.dev/docs/mcp\n\n## About\n\nHumanHours is the ROI meter for AI agents: it measures the hours and money your agents\nsave, and enriches companies with real wage data so every saving is grounded in what the\nwork actually costs. Learn more at https://humanhours.dev.\n",
  "bytes": 2983,
  "sha": "9a0cbaa8d9d37e7cbf0b2e1e0e22febc06edc5a7aa1dd7867ff6a21a90c0c69d",
  "repo_slug": "triadgit/humanhours-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_dev_humanhours_humanhours_c3460975/readme"
}