{"slug":"ai-builder-club-agent-context-audit","source_name":"ai-builder-club/agent-context-audit","name":"AI Builder Club/Agent Context Audit","description":"Audit a repo's agent context — CLAUDE.md files, codebase docs, skills, and tool/MCP designs — against Anthropic's Claude 5 context-engineering guidance (\"unhobbling\": Anthropic cut ~80% of Claude Code's system prompt with no eval loss). Finds overconstraint, conflicting instructions, redundancy, stale facts, and missing \"unknown knowns\"; produces a scored findings report with concrete rewrites, then applies approved fixes. Use when someone says \"audit my CLAUDE.md\", \"context audit\", \"unhobble th","version":1,"lift":{"pass_rate_delta_pts":13.64,"pass_rate_pct":81.8,"total_cases":22,"passed_cases":18,"tokens_delta_pct":88.7,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-15T20:56:39.779859+00:00"},"skill_score":0.8182,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":13.64,"with_pass_pct":81.8,"without_pass_pct":68.2,"tokens_delta_pct":88.7,"turns_delta_pct":0,"total_cases":22,"cases_aggregated":20,"verdict":"mixed","never_hurt":false,"completed_at":"2026-08-15T20:56:39.779859+00:00","run_id":"4070b177-5dc2-4fb7-adb1-e1d9e6e18889","version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":null,"install_count":0,"manifest_hash":"e577806d9e8b57a0a4c701be782e4b02331cada75b458938ddedb6ccc97e3dad","raw_url":"https://app.decimal.ai/s/ai-builder-club-agent-context-audit/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/ai-builder-club-agent-context-audit"}