Catalog PyPI
PACKAGE EXPLORERView in registry

agentic-audit-mcp

PyPI

MCP server for AgenticAudit — lets AI agents query their own compliance audit trail

aiauditcompliancemcpmodel-context-protocol
Versions collected1
Direct dependencies2From release 0.3.0
Last observed20 Oct, 22:12 UTC

README

Version 0.3.0 documentation

AgenticAudit MCP Server

MCP server that lets AI agents query their own AgenticAudit compliance trail.

Agents can review their recent actions, check risk summaries, dry-run risk checks, and query AI Act compliance status.

Install

pip install agentic-audit-mcp

Or run directly:

uvx agentic-audit-mcp

Configure in Claude Code

// ~/.claude/settings.json
{
  "mcpServers": {
    "agentaudit": {
      "command": "uvx",
      "args": ["agentic-audit-mcp"],
      "env": {
        "AGENTAUDIT_API_KEY": "aa_live_xxxxx",
        "AGENTAUDIT_BASE_URL": "http://localhost:8000"
      }
    }
  }
}

Tools

ToolDescription
get_my_audit_eventsReview recent actions and risk levels
get_session_risk_summaryRisk breakdown for the current session
check_action_riskDry-run risk check without logging
list_ai_systemsList registered AI systems with compliance status
get_compliance_statusAI Act compliance score and check results
suggest_classificationSuggest risk classification from event patterns

License

Apache 2.0

Downloads & popularity

Awaiting observations

No metric observations yet

Download counts and popularity will appear here as the collector receives them.

Versions 1

Collected releases
VersionPublishedDirect dependenciesOpen version
0.3.025 Mar 20262

Package advisories

0
No matching advisories collected

This describes our collected data; it is not a security assessment.

Repository

PENDING
dorianganessa/agentic-audit

Shared repository data · GitHub

License detected in repository
Not collected

Metadata history

Meaningful changes
Current collected state Current

Observed since 20 Oct, 22:12 UTC

No superseded metadata states collected. Unchanged observations do not create new revisions.

Record identity

package:d64f4dc80976ff49efc75a7b1b074637a312f183a8568765281931664a6fa9ee