{
  "markdown": "# Taste\n\n[![smithery badge](https://smithery.ai/badge/with0utwhy/taste-mcp)](https://smithery.ai/servers/with0utwhy/taste-mcp)\n\nExpert review for AI agents. On-chain proof of human review.\n\nThis is the public manifest repository for the Taste MCP server. The server is\nhosted — there is nothing to install. `server.json` is the discovery manifest\nconsumed by MCP registries; the server source is maintained privately.\n\n## what\n\nPay-per-call human work for AI agents over MCP + x402. Vetted humans review an\nAI agent's output across any media (text, images, video, audio, social posts),\naudit reasoning chains, run prepublish safety checks, gate high-stakes actions\nfor human approval, and convene curated expert teams for working sessions on\nideas, problems, or decisions. Approved outputs receive a Taste content\ncertificate recorded on Base — any downstream agent can verify the review.\n\n## connect\n\n```\nMCP endpoint:  https://api.humantaste.app/mcp/\nTransport:     Streamable HTTP (MCP 2025-06-18)\nPayment:       x402, USDC on Base mainnet, $1.00–$15.00 per paid call\nFree tools:    list_offerings, list_expert_profiles, get_result, verify_certificate\nTypical wait:  5–30 min for evaluation calls; up to 24h for a think tank session\n```\n\nNo authentication, no API key. Paid tools answer an unauthenticated call with\nHTTP 402; sign an EIP-3009 USDC transfer and retry.\n\n## tools\n\nFree (no payment, no session):\n\n| Tool | Purpose |\n|---|---|\n| `list_offerings` | list offerings, prices, certificate flag, operating-hours state |\n| `list_expert_profiles` | aggregate availability + estimated response time, no PII |\n| `get_result` | poll session status by sessionId or order reference code; returns the full deliverable when complete |\n| `verify_certificate` | read-only Taste content certificate lookup on Base, rate-limited 60/min/IP |\n\nPaid (x402 — USDC on Base):\n\n| Tool | Price | Purpose |\n|---|---|---|\n| `verify_external_source` | $1.00 | verify whether an external source, project, or claim is trustworthy |\n| `review_content` | $1.00 | review AI-generated content for hallucinations, factual errors, domain mistakes — issues a certificate |\n| `verify_reasoning_chain` | $1.00 | audit a step-by-step reasoning chain for mid-chain errors — issues a certificate |\n| `prepublish_review` | $1.00 | review content for cultural sensitivity, brand safety, derivative risk before publishing — issues a certificate |\n| `predict_audience_reaction` | $1.00 | a human matching a target demographic rates and reacts to content |\n| `arbitrate_dispute` | $1.00 | an impartial human arbiter decides whether a deliverable meets a contract |\n| `request_human_approval` | $1.00 | pause a workflow for explicit human approval before an irreversible action |\n| `order_think_tank_session_30` | $5.00 | convene a curated team of human experts for a 30-min working session on an idea, problem, or decision; returns synthesized directions (1 free revision turn) |\n| `order_think_tank_session_60` | $15.00 | as above, 60-min session (2 free revision turns) |\n\nEach think tank order returns a `TASTE-XXXX-XXXX-XXXX` reference code. Two\ncontextual helper tools — `request_think_tank_revision` and\n`request_illustration_revision` — spend a free revision turn against that code.\n\nThe live tool list is authoritative — `server.json` sets `tools: \"dynamic\"`,\nso MCP clients call `tools/list` against the endpoint for the current surface.\n\n## example\n\n```bash\ncurl -X POST https://api.humantaste.app/mcp/ \\\n  -H \"Content-Type: application/json\" \\\n  -H \"Accept: application/json, text/event-stream\" \\\n  -d '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/call\",\n       \"params\":{\"name\":\"list_offerings\",\"arguments\":{}}}'\n```\n\nPaid calls return HTTP 402 with an `X-Payment-Requirements` header (base64\nJSON). Sign EIP-3009 `transferWithAuthorization` for USDC on Base, then retry\nthe same request with an `X-Payment` header.\n\n## compliance\n\nEU AI Act Article 50(2) exempts AI-generated content from mandatory\nAI-disclosure labels when it has undergone human review or editorial control.\nTaste content certificates are the on-chain audit trail for that exemption.\n\n## links\n\n- One-pager for agents: https://api.humantaste.app/llms.txt\n- Full integration reference: https://api.humantaste.app/llms-full.txt\n- MCP Server Card: https://api.humantaste.app/.well-known/mcp.json\n- Paid-resources catalog: https://api.humantaste.app/.well-known/x402\n- Brand: https://humantaste.app\n",
  "bytes": 4439,
  "sha": "b60aaf9a02cede9a97064fcfe4f63e553f6b1ab3a0f92de9f4fdc85e5f12619a",
  "repo_slug": "with0utwhy/taste-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_app_humantaste_taste_mcp_da697849/readme"
}