{
  "markdown": "<p align=\"center\">\n  <img src=\"assets/logo.png\" width=\"80\" alt=\"AgentPay\">\n</p>\n\n# AgentAudit MCP\n\nYour agent now has an immutable, tamper-evident log of every action it takes — each event cryptographically linked to the previous one so any modification breaks the chain and is immediately detectable.\n\n## What your agent can do\n\n- Log any of 16 event types (payments, tasks, disputes, identity checks, messages) with a SHA-256 hash that includes the previous event's hash\n- Retrieve the full history of any agent by ID — every action, in order, with timestamps\n- Search across the audit trail by agent, event type, severity level, or date range\n- Verify chain integrity between any two events and get a per-event pass/fail breakdown showing exactly which link was tampered\n- Track five severity levels — `info`, `notice`, `warning`, `error`, `critical` — for compliance triage\n- Get statistics across the entire audit trail: events by type, by severity, and most active agents\n\n## Installation\n\n**Requires:** Python 3.10+, `mcp` and `fastmcp` packages.\n\n```bash\npip install mcp fastmcp\n```\n\n**Claude Desktop** — add to `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS) or `%APPDATA%\\Claude\\claude_desktop_config.json` (Windows):\n\n```json\n{\n  \"mcpServers\": {\n    \"agent-audit\": {\n      \"command\": \"python\",\n      \"args\": [\"/absolute/path/to/agent-audit-mcp/server.py\"]\n    }\n  }\n}\n```\n\n**Cursor** — add to `.cursor/mcp.json` in your project root:\n\n```json\n{\n  \"mcpServers\": {\n    \"agent-audit\": {\n      \"command\": \"python\",\n      \"args\": [\"/absolute/path/to/agent-audit-mcp/server.py\"]\n    }\n  }\n}\n```\n\nData is stored locally at `~/.agentaudit/chain.json`. No data leaves your machine.\n\n## Tool Reference\n\n| Tool | Description | Key params |\n|------|-------------|------------|\n| `audit_log` | Log an event and append it to the hash chain | `event_type` (one of 16), `agent_id`, `severity`, `details_json` |\n| `audit_get_event` | Retrieve full event details by ID | `event_id` |\n| `audit_search` | Filter events by agent, type, severity, or date range | `agent_id`, `event_type`, `severity`, `from_date`, `to_date`, `max_results` |\n| `audit_get_agent_history` | All events for one agent, newest first | `agent_id`, `max_events` |\n| `audit_verify_chain` | Check SHA-256 chain integrity between two events; returns per-event PASS/FAIL | `from_event_id`, `to_event_id` |\n| `audit_stats` | Summary: total events, breakdown by type/severity, top 5 active agents | — |\n\n**Supported event types:** `task_created`, `task_assigned`, `task_completed`, `task_failed`, `payment_sent`, `payment_received`, `payment_refunded`, `dispute_raised`, `dispute_resolved`, `identity_verified`, `capability_used`, `agent_registered`, `agent_rated`, `agent_offline`, `message_sent`, `message_received`\n\n## Security\n\nAudit data is written atomically (write-to-temp, then `os.replace`) to prevent partial writes from corrupting the chain. The hash covers every field including the previous event's hash — you cannot insert, delete, or modify any event without breaking all subsequent hashes. `audit_verify_chain` recomputes each hash from scratch rather than trusting stored values.\n\nAll data is stored locally. No network requests are made by this server.\n\n## Pricing\n\n| Plan | Price | Included |\n|------|-------|----------|\n| Pro | $19/month | Unlimited audit events, full chain verification |\n\n[Subscribe via Stripe](https://buy.stripe.com/dRm6oJ4Hd2Jugek0wz1oI0m)\n\n## License\n\nProprietary — see subscription terms. Source: [github.com/Rumblingb/agent-audit-mcp](https://github.com/Rumblingb/agent-audit-mcp)\n",
  "bytes": 3610,
  "sha": "07cb026909306438075a88aa67cafefd5f58515ffb8d58a09cfb9db97a8ca50a",
  "repo_slug": "rumblingb/agent-audit-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_rumblingb_agent_audit_mcp_c26a6623/readme"
}