qingxuantang/tar-engine
Audit AI skill safety before you ship. Static + semantic + adversarial audit pipeline for SKILL.md, Codex skill.yaml, Claude Code commands, and OpenCode. Free hosted tier on tarai.dev.
Audit AI skill safety before you ship. Static + semantic + adversarial audit pipeline for SKILL.md, Codex skill.yaml, Claude Code commands, and OpenCode. Free hosted tier on tarai.dev.
npx add-skill qingxuantang/tar-engineAudit AI skill safety before you ship. Static + semantic + adversarial audit pipeline for SKILL.md, Codex skill.yaml, Claude Code commands, and OpenCode. Free hosted tier on tarai.dev.
🏆 AI Agent 生态排行榜：Skills · MCP 服务器 · Prompt 库 · 框架 · 深度研究，按 GitHub Stars 每日更新 | Leaderboard for AI agent ecosystem — Skills, MCP, Prompts, Frameworks & Research.
OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards
Claude/Codex skill and local workflow layer for efficient interaction with Lean 4 and Monte-Carlo Tree Search.
Install academic research tools and paper search capabilities across your AI coding agents with a single command.
Agent skills for developing a Rill project
Catches AI prose before it ships. Density-based slop detector for Claude Code: 45 patterns + 150 vocab tells + 33 structure tells, calibrated against PubMed/arXiv data, with model fingerprinting for GPT/Claude/Gemini. Passes its own test.