The EU AI Act becomes enforceable in August 2026. If your AI agents make decisions that affect people - hiring, lending, healthcare, legal - you need audit trails.

Most AI agent frameworks have no built-in governance. LangChain, CrewAI, and OpenAI Agents SDK all let you build powerful agents, but none of them log what the agent did, why it did it, or whether a human approved it.

The requirement

Articles 9-15 of the EU AI Act require:

  • Automatic logging of agent actions (Article 12)
  • Human oversight mechanisms (Article 14)
  • Risk management documentation (Article 9)
  • Technical documentation of system behavior (Article 11)

The solution

asqav adds governance to any Python AI agent in one decorator:

from asqav import audit

@audit
def my_agent_call(prompt):
    return llm.invoke(prompt)

Every call gets a tamper-evident audit trail with:

  • Input/output recording
  • Timestamp and caller identity
  • Quantum-safe digital signature (ML-DSA / FIPS 204)
  • Policy evaluation results

Works with your stack

Install

pip install asqav

Open source, MIT licensed, no vendor lock-in.