Aegisora v0.1.0 - The Open-Source Launch 🚀

🚀 Initial Open-Source Release

Welcome to the first official release of Aegisora — the zero-trust runtime security and governance layer for autonomous AI agents.

✨ Core Capabilities

• The Zero-Latency Proxy: Intercept and validate agent tool calls in real-time (< 10ms) before they touch production.
• Asynchronous Human Escalation: Solving the "Binary Trap." High-risk, ambiguous actions are automatically paused and routed to a Human Review Queue, allowing human-in-the-loop governance without breaking deterministic workflows.
• Seamless Integrations: Drop-in middleware support for LangChain, LangGraph, and standard LLM orchestrations.
• Enterprise Audit Logs: Full observability and execution integrity proofs backed by Supabase.

🛠 Getting Started

Check out the README.md for our quickstart guide, and visit the examples/ directory to see Aegisora running inside a stateful LangGraph workflow.