agent-audit-logger-mcp

Name: agent-audit-logger-mcp · Audit Logging MCP for AI agents · MEOK AI Labs
Author: MEOK AI Labs

Hash-chained HMAC-signed audit log MCP for A2A (agent-to-agent) calls. Every tool-call, agent-handoff, decision gets a t

Live MCP server PyPI v1.1.6 MIT

RegulationAudit Logging

Latest version1.1.6

AuthorMEOK AI Labs

LicenseMIT

What it does

Hash-chained HMAC-signed audit log MCP for A2A (agent-to-agent) calls. Every tool-call, agent-handoff, decision gets a t

Install

pip install agent-audit-logger-mcp
# or
uv pip install agent-audit-logger-mcp

Use with your AI agent

Add the following to your MCP client configuration (Claude Desktop, Cursor, Cline, or any MCP-compatible host):

# Add to your MCP client config (e.g. claude_desktop_config.json):
{
  "mcpServers": {
    "agent-audit-logger-mcp": {
      "command": "uvx",
      "args": ["agent-audit-logger-mcp"]
    }
  }
}

How it works

1. Standards-aligned

References the actual regulation text from Audit Logging — not a paraphrase. When you ask about Article 9, you get Article 9.

2. Deterministic

Same input → same output. Suitable for compliance evidence packs, audits, and regulator review.

3. Agent-native

Tools are first-class. Your agent can call them by name, pass arguments, and use results in subsequent steps.

4. Audit-ready

Every call is hash-chained, signed, and logged. Integrates with agent-audit-logger-mcp for tamper-evident audit trails.

Author & provenance

Built by MEOK AI Labs (CSOAI Ltd, UK Companies House 16939677).
Source: https://github.com/CSOAI-ORG/agent-audit-logger-mcp
PyPI: pypi.org/project/agent-audit-logger-mcp

Compliance disclaimer

MEOK MCPs are tooling for engineering and compliance teams. They do not constitute legal advice. For final compliance decisions, consult qualified counsel in your jurisdiction.