AgenticAudit¶
The missing audit layer for AI agents.
Every action your AI agents take — logged, classified, and audit-ready. From Claude Code to Cowork.
-
Get started in 5 minutes
Start AgenticAudit with Docker Compose, log your first event, and see it classified.
-
Integrate with your agents
Claude Code hooks, Cowork plugin, LangChain callback, REST API — pick your path.
-
Understand the concepts
How PII detection, risk scoring, and compliance framework mapping work.
-
API Reference
Every endpoint documented. Request/response schemas, error codes, examples.
What it does¶
Your agent does something → AgenticAudit captures it → classifies risk and detects PII → maps to GDPR, AI Act, and SOC 2 articles. Automatically.
from agentaudit import AgentAudit
audit = AgentAudit(api_key="aa_live_xxxxx")
event = audit.log(
agent_id="booking-agent-v2",
action="access_customer_record",
data={"customer_email": "[email protected]"},
reasoning="Customer requested booking modification",
)
print(event.risk_level) # "medium"
print(event.pii_detected) # True
print(event.frameworks) # {"gdpr": ["art_30"], "ai_act": ["art_14"]}
Works with¶
| Agent | Support | Method |
|---|---|---|
| Claude Code |  Full | Deterministic hooks, enterprise-enforceable |
| Cowork | Full | Plugin with hooks, marketplace deployment |
| LangChain | Full | Callback handler |
| CrewAI | Full | Event hook |
| Codex | Partial | Transcript parsing |
| Any agent | Full | REST API / Python SDK |