Turn Chaos Into Structure. A Type-Safe AI Agent that extracts valid JSON from unstructured data using PydanticAI, FastHTML, and Gemini 2.5.

Interactive Phoenix LiveView demonstrations of the Crucible Framework - showcasing ensemble voting, request hedging, statistical analysis, and more with mock LLMs

Reference implementation of CAAF — three-pillar agent framework with monotonic convergence.

Reliability and audit-evidence testing for LLM agents - wrap any agent, assert behavior, measure determinism, check grounding, emit an audit-grade report.

Official implementation of CHARM: Cascading Hallucination Aware Resolution and Mitigation for multi-step agentic RAG pipelines.

Map where your bolted-on AI feature breaks before customers do. A free Claude Code tool: fragility map, reliability score across six dimensions, ranked gaps, and a 30-day plan. Built by a threat-intel practitioner.

CrucibleFramework: A scientific platform for LLM reliability research on the BEAM

Reliability and hallucination mitigation research for tool-augmented legal AI agents using QC-Sentinel verification architecture.

Collection of LLM failure modes used on failmodes.com

Public artifact bundle for the preprint 'Lightweight Evaluation and Operational Scorecards for Tool-Using AI Agents'

Production-style LLM evaluation harness for structured clinical extraction — compares prompt strategies across accuracy, cost, and hallucination.

AI agent metacognition skills distilled from David Dunning's Self-Insight (2005): calibration, Dunning–Kruger awareness, the outside view, and feedback discipline for Claude Code, Codex, and compatible AI agents.

Preprint paper package — Lightweight Evaluation and Operational Scorecards for Tool-Using AI Agents (Zenodo DOI 10.5281/zenodo.20034550)