4 ms·
Show HN: I tested 11 AI frameworks for basic security – none passed
- Acacian 6mo agoLangChain, CrewAI, OpenAI Agents, Anthropic, LiteLLM, Pydantic AI, Google ADK — I went through 11 frameworks looking for basic runtime security: injection detection, PII masking, audit trails. None of them had it. So I built a monkey-patching layer that intercepts LLM calls and runs them through guardrails: import aegis; aegis.init() Patches whatever frameworks you have installed. ~2.6ms overhead. The nastiest find: streaming responses skip middleware entirely. Content leaks before any check runs. I wrote a streaming engine that auto-selects between windowed scanning and full buffering depending on what the guardrail needs — PII like "078-05-1120" can split across chunks, so regex won't catch it without the full buffer. Context: https://github.com/langchain-ai/langchain/issues/35011 https://github.com/langchain-ai/langchain/issues/35011 Source: https://github.com/Acacian/aegis https://github.com/Acacian/aegis
- hyperlambda 6mo ago[dead]