joergmichno

5 ms·

1. 1.▲
   by joergmichno 7mo ago

   Fair point — and one I thought about carefully before open-sourcing. A few reasons why I think open patterns are actually the right call: 1. The patterns are already public. Most prompt injection techniques are documented on GitHub, in rese
2. 2.▲
   by joergmichno 7mo ago

   A bit more context on why pattern matching instead of ML: 1. Speed: <10ms vs 200-500ms for LLM-based checks means you can scan every user message without adding latency. 2. Cost: No API calls to OpenAI/Anthropic for detection = pred
3. 3.▲
   Show HN: ClawGuard – Detect 42 prompt injection patterns in <10ms(github.com)

   2 points by joergmichno 7mo ago|3 comments