Ai Engineering Toolkit

Six production-grade workflows that encode repeatable AI engineering methodologies with quantitative scoring. The prompt evaluator uses 8-dimension scoring (0-100 scale) with auto-rewrites for weak spots. Context budget planner catches output zone squeeze before truncation hits. RAG architect walks the full decision tree from chunking to retrieval patterns. Agent safety guard runs 65-point red-team audits across injection, extraction, and tool abuse vectors (requires sandboxing, marked offensive risk). Eval harness builder ships CI-ready metric systems. Product sense coach structures pre-code strategic thinking. This is for teams that want baseline scoring systems and security checklists baked into workflows, not ad-hoc prompting.

Install to Claude Code

npx -y skills add sickn33/antigravity-awesome-skills --skill ai-engineering-toolkit --agent claude-code