Just submitted asqav to Hacker News: https://news.ycombinator.com/item?id=47665415
asqav is an open-source Python SDK that signs every AI agent action with ML-DSA-65 (quantum-safe, FIPS 204). Three lines to add tamper-evident audit trails to any agent framework.
pip install asqav
import asqav
asqav.init(api_key=\"sk_...\")
agent = asqav.Agent.create(\"my-agent\")
agent.sign(\"api:call\", {\"model\": \"gpt-4\"})
Works with LangChain, CrewAI, OpenAI Agents, Haystack, LiteLLM. MCP server for Claude Desktop. MIT licensed.
We built this because the EU AI Act Article 12 requires tamper-evident logging by August 2026, and nothing existed that was both cryptographically serious and easy to integrate.
GitHub: https://github.com/jagmarques/asqav-sdk
Docs: https://asqav.com/docs/
MCP server: https://github.com/jagmarques/asqav-mcp
Compliance scanner: https://github.com/jagmarques/asqav-compliance
If you find it useful, a star on GitHub helps others find it.
Top comments (1)
One surprising insight from our work with enterprise teams is that while many focus on building AI agents, it's often the audit trails that become the most critical aspect. Developers frequently overlook how these trails facilitate debugging and compliance, especially in quantum-safe environments. Building agents is essential, but integrating robust audit mechanisms can be just as vital for operational success. - Ali Muwwakkil (ali-muwwakkil on LinkedIn)