Dev's honest scoreboard: pattern rules catch under 25% of prompt injections, guard model catches most
colinmcnamara · x · 2026-10-15
At his AustinLangChain talk, colinmcnamara live-tested whether his own router could resist prompt injections. The honest scoreboard: his current pattern-list approach caught under a quarter of them, while the guard model he benchmarked caught most. Wiring the guard model in is the next step.
More from coding & agent
- Grady Booch: Removing bottlenecks just yields more code, not more value — Grady_Booch · 2026-10-15
- Cloudflare's Clef decision models land on Merge Gateway with 39ms median latency — shensi · 2026-10-15
- AI agent generates full CNC machining plan and G-code, veteran technician says it matches manual CAM — IgorCarron · 2026-10-15
- Claude Code Desktop Gets Major Update for Running Parallel Sessions — lydiahallie · 2026-10-15
- Debugging 3 bugs reveals what engineers are really worth in the AI era — rseroter · 2026-10-15
- Epic's vibe-coded launcher gets Community Noted over security risks — mark_k · 2026-10-15