Pre-execution semantic firewall for AI agents. Signed policy verdicts, BYOK, zero-trust receipts.
python typescript mcp ed25519 compliance ai-agents ai-security github-actions byok langchain prompt-injection llm-security eu-ai-act pydantic-ai agent-security owasp-llm semantic-firewall microsoft-agt tool-call-interception cryptographic-receipt
-
Updated
Jul 21, 2026 - Python