Pinned Loading

1. llm-credit-decision-audit llm-credit-decision-audit Public

   Does an AI loan agent's stated denial reason match the reason it actually acted on? A causal audit harness that repairs the cited factor and replays the application to find out. LLM evaluation for …

   Python

2. rag-citation-guardrail rag-citation-guardrail Public

   Every RAG eval reports faithfulness. None reports over-refusal beside it. An eval harness and policy engine for refusal correctness in regulated RAG: synthetic KYC/AML corpus, deterministic verifie…

   Python

3. agent-memory-staleness-audit agent-memory-staleness-audit Public

   Your agent still thinks they work at Acme, and nothing in memory contradicts it. Scores agent memories for silent staleness from provenance, fact-type volatility, and supersession without contradic…

   Python

4. Project-Chronos Project-Chronos Public

   TypeScript

5. agent-trace-to-evals agent-trace-to-evals Public

   You have thousands of agent traces and no evals. Mines production traces for tool-call invariants, proposes them as pytest assertions with evidence counts, and gates CI on the ones a human approved…

   Python

6. open-weight-ai-agent-game-voice-generator open-weight-ai-agent-game-voice-generator Public

   Open-weight AI voice generation system for game characters — 31 Colab modules on Qwen3-TTS covering NPC dialogue packs, branching dialogue trees, boss encounters, emotion grids, and lifetime voice …

   Jupyter Notebook