Pinned Loading

1. AIObjectives/tttc-light-jsAIObjectives/tttc-light-jsPublic

   An open-source LLM analysis tool to improve collective discourse and decision-making.

   TypeScript 56 16

2. headroomheadroomPublic

   More human oversight can make an AI agent less safe. Headroom is a human-in-the-loop firewall for coding agents that measures when to trust the human: oversight as resource allocation, not just cla…

   Python 2

3. OperatorLockOperatorLockPublic

   Behavioral risk-control and execution infrastructure that enforces trading discipline through hard server-side gates, temporal locks, and cooldowns.

   HTML

4. agent-governoragent-governorPublic

   Reusable oversight & evaluation layer for autonomous AI agents: gate each action (auto-send / escalate / hold), measured on held-out data with an LLM-as-judge, and self-improving under evaluation. …

   Python

5. grounded-agentgrounded-agentPublic

   Voice/text Q&A grounded in your documents - every answer cited, and it escalates to a human instead of guessing. RAG on Moss + LiveKit voice + Claude, correctly layered.

   Python

6. AIObjectives/elicitation_botAIObjectives/elicitation_botPublic

   Python 2 2