🔍️ Extract provenance information (W3C PROV) from GitLab projects.
A provenance library for bioinformatics workflows 🧬 🔀 📝
⬆️ Import W3C PROV documents to Neo4j.
A reference implementation for the BACARDI PROV data model.
Post-hoc trace reconstruction for AI agents: turns raw runtime traces into per-property reconstructability reports naming evidenced, missing, and opaque decision facts.
Reference implementation, full provenance dataset (7,004 experiments), and revision-ready submission package for the Nature Communications study 'Same Prompt, Different Answer: Hidden Non-Determinism in LLM APIs Undermines Scientific Reproducibility' (Rover et al., 2026, under Major Revision).
PAC-AI: Protocol for Auditable Context in AI — Python SDK
PAC-AI protocol use cases — Healthcare (Art. 14 human oversight) and Education (Art. 13 negative proof) compliance scenarios with benchmarks, built on the jhcontext SDK for EU AI Act auditability.
AWS production deployment of the PAC-AI protocol (jhcontext) — Chalice REST API + MCP proxy on Lambda, DynamoDB + S3 storage, and CrewAI multi-agent scenarios demonstrating EU AI Act compliance (Articles 13 & 14) with auditable context envelopes and W3C PROV provenance.
Add a description, image, and links to the w3c-prov topic page so that developers can more easily learn about it.
To associate your repository with the w3c-prov topic, visit your repo's landing page and select "manage topics."