BitsExplorer

BitsExplorer finds what people are building with AI, every day. See today's picks →

Preview of Llm Redteam Guardrail Audit

Llm Redteam Guardrail Audit

Red-team and guardrail audit of a RAG support chatbot: 59 attacks incl. indirect prompt injection via poisoned articles, canary leak detection, LLM judge, 3 defense configs, utility check and controls.

Description excerpted from the original listing, which is linked below.

Open it Source code

The week in AI-built software

Every Monday: how many projects were built with AI that week, which tools built them, where they came from, and the week's picks.

Read the latest week