Pinned Loading

1. explain-simply explain-simply Public

   Plain-language Agent Skill for decision-ready technical reports in Claude Code and Codex

   1

2. skill-vault skill-vault Public

   Audit Agent Skill context overhead in OpenAI Codex and Claude Code; reversibly lazy-load approved Codex skills without changing triggers, with safe rollback.

   JavaScript 1

3. ai-safe-driver ai-safe-driver Public

   AI agent drift detection and recovery for Claude Code and OpenAI Codex—stop repeated tool-call, instruction-following, and output-format failures with permission-gated handovers.

   JavaScript 1