What it does
This MCP server keeps a hash-chained, HMAC-signed audit log of agent-to-agent (A2A) calls. Each tool call, agent handoff and decision is stored as a tamper-evident signed record. It exposes three tools: log, which writes an entry; verify_chain, which checks the integrity of the chain; and search, which queries stored records.
Teams that need a traceable record of multi-agent activity would reach for it. The publisher links the records to EU AI Act Article 12 automatic logs, DORA Article 17 ICT incident logs and ISO 42001 clause 9 monitoring, and offers compliance reporting. Signed end-of-day attestations are part of a paid tier.
It runs over stdio transport. It can be installed from PyPI and launched as a command, or added to the configuration of an MCP client such as Claude Desktop. Once configured, a user can call the log, verify_chain and search tools from their assistant.
Limitations and requirements
The free tier is limited to 1,000 log entries per day and its logs are ephemeral. Unlimited logging, signed end-of-day attestations and chain integrity reports require the paid Pro tier. The publisher lists SIEM webhook push and a retention policy for the Enterprise tier.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.CSOAI-ORG/agent-audit-logger-mcp
Hash-chained HMAC-signed audit log MCP for A2A (agent-to-agent) calls. Every tool-call, agent-ha...
v1.1.10As shown on registry.modelcontextprotocol.io when we last read it