Audit a skill by running a paired probe — the same task once with the skill loaded and once without — segment both traces into goal-directed phases, align phases, and emit a SIP report (surface anchoring, template copy, excess planning, task recovery, off-task artifact). Use whenever a skill is created, modified, or proposed for retirement. Pass-rate is BLIND to most skill effects: CTA (arxiv 2605.11946v1) shows a single skill can produce 522 measurable behavioural changes across 49 tasks while pass-rate moves only +0.3%. Triggers: "audit this skill", "is this skill helping", "retire skill", "before shipping skill", "behavioural impact of skill X", or any skill review event.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
When you ship a skill, pass-rate is the easy metric — and it's blind to almost everything that matters. CTA (arxiv 2605.11946v1) shows a single skill can produce 522 measurable behavioural changes across 49 tasks while pass-rate moves only +0.3%. The interesting effects (surface anchoring on the skill's vocabulary, template copy without adaptation, excess planning loops, recovery from off-task drift, generation of off-task artifacts) live entirely below pass-rate.
The remedy is a paired-trace audit: run the same probe task twice, with and without the skill, and compare the behaviours directly rather than the headline metric.
For each skill under audit:
.planning/patterns/skill-audits/<skill>-<timestamp>.md with phase comparison table + SIP distribution + retire/refine/keep recommendation.| Dominant SIP | Verdict |
|---|---|
| task-recovery + correct artifacts | keep (genuine value) |
| surface-anchoring with no task-recovery | refine (skill is leaking style, not capability) |
| template-copy on most tasks | refine (lower template specificity) |
| off-task-artifact on >25% of probes | retire (skill is causing distraction) |
| excess-planning with no completion delta | refine (skill adds ceremony) |
loop, schedule) — audit dominant SIPs would always be "task-recovery" trivially.done-retirement — required gate before retiring a skill.gsd-cleanup — sweeps audits on a milestone-close cadence.simplify, code-review — those audit code; this audits skill behaviour.agent-systems / integration-evaluation / agent-paired-trace-auditskill-integration (manages the skill lifecycle that this audits)まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Provides web accessibility best practices for semantic HTML, ARIA, keyboard navigation, color contrast, and screen reader patterns. Use when building UI components, reviewing accessibility, or when user mentions 'a11y', 'accessibility', 'ARIA', 'screen reader', 'keyboard navigation', 'WCAG'.
日本語の概要は準備中です。原文の説明を表示しています。
Active listening techniques for effective communication. Covers attending behaviors, paraphrasing, reflective listening, clarifying questions, empathic response, barriers to listening, listening in conflict, and cross-cultural listening. Use when building listening skills, improving understanding in conversation, mediating disputes, or analyzing communication breakdowns.
日本語の概要は準備中です。原文の説明を表示しています。
Adversarial spec-compliance PR review — cross-references diffs against approved specs, verifies runtime claims against source, detects competing PRs, audits scope/convention compliance. Use before merging.
日本語の概要は準備中です。原文の説明を表示しています。
Provides best practices for AI agent orchestration including MCP servers, A2A protocol, multi-agent coordination, and swarm architectures. Use when designing agent systems, configuring MCP servers, setting up agent teams, or when user mentions 'MCP', 'A2A', 'agent orchestration', 'multi-agent', 'swarm', 'agent team', 'LangGraph', 'CrewAI', 'AutoGen'.
日本語の概要は準備中です。原文の説明を表示しています。
Symbolic manipulation, equation solving, and algebraic structures for mathematical reasoning. Covers distributive law, factoring, completing the square, linear through polynomial equation solving, systems of equations (substitution, elimination, Gaussian elimination, matrix methods), algebraic structures (groups, rings, fields), modular arithmetic, polynomial theory, and inequalities. Use when solving equations, simplifying expressions, working with algebraic structures, or performing symbolic manipulation.
日本語の概要は準備中です。原文の説明を表示しています。
Understanding how algorithmic systems shape what users see, know, and do -- from recommendation feeds to search ranking to credit scoring to hiring software. Covers the mechanics of recommendation systems, algorithmic bias and its sources, personalization's effects on information diets, opacity and accountability, AI limitations (hallucination, confident wrongness), and the human-in-the-loop question. Use when a learner needs to think critically about why particular content reached them.
日本語の概要は準備中です。原文の説明を表示しています。