Post-hoc trace reconstruction for AI agents: turns raw runtime traces into per-property reconstructability reports naming evidenced, missing, and opaque decision facts.
-
Updated
Apr 29, 2026 - Python
Post-hoc trace reconstruction for AI agents: turns raw runtime traces into per-property reconstructability reports naming evidenced, missing, and opaque decision facts.
Autonomous multi-game AI platform where agents reason, adapt, and compete in real time.
Enable multi-game play with AI as an autonomous player using live game states, supporting diverse genres and revealing decision traces for better AI interaction.
🛠️ Log and trace your AI's decisions securely with DecisionTrace, ensuring transparency and integrity in every choice it makes.
A minimal append-only ledger for recording and replaying decision processes (Event → Signal → Decision → Action).
Semantic decision telemetry integration for CrewAI.
Reproducibility package for an arXiv pilot measuring property-level reconstructability of agent decisions across vendor SDK adapter regimes.
DSL-driven decision orchestration engine with traceable execution, human-in-the-loop, and boundary control.
JSON schema for DecisionTrace v0.1 — minimal, portable event record for governed AI decisions; includes examples.
Add a description, image, and links to the decision-trace topic page so that developers can more easily learn about it.
To associate your repository with the decision-trace topic, visit your repo's landing page and select "manage topics."