Pinned Loading

1. tracefork tracefork Public

   Time-travel debugger for AI agents: record a run to a content-addressed tape, replay it bit-exact for $0 (hash-verified), fork any step, and measure causal blame with confidence intervals.

   Python 1

2. rag-receipts rag-receipts Public

   Adaptive agentic RAG engine with a built-in Ablation Lab: hybrid retrieval (BM25 + dense + RRF) -> Cohere rerank -> System-1/System-2 router, with reproducible receipts measuring each component's c…

   Python 1

3. faithfulnessbench faithfulnessbench Public

   A causal-intervention harness for measuring — and validating — chain-of-thought faithfulness in reasoning models.

   Python 1

4. evevator-system evevator-system Public

   JavaScript