Research industry approaches to practices discussed in the conversation and assess their fit for the decisions at hand.
The English summary is being prepared. Showing the original description.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
Extract the generalizable practices behind the conversation's proposals and decisions, then research established approaches and relevant current trends on the web. Prioritize the topic in $ARGUMENTS when provided.
Frame questions around the decisions the user needs to make; let their scope determine the breadth and depth of research. If neither the conversation nor $ARGUMENTS provides a usable topic, ask for one.
Prefer official documentation and primary sources. Cross-check claims about common practice or consensus against independent sources, and keep individual examples distinct from evidence of broad adoption. Note publication or update dates where available; use recent evidence for trends and check versions when guidance depends on them.
Finish when the evidence supports the relevant comparisons and recommendations, or further searching yields no material evidence within the requested scope. Report unresolved conflicts and gaps rather than expanding into unrelated topics or presenting uncertainty as consensus.
Explain the approaches found, how they compare with the conversation's proposals, and whether to adopt, adapt, or skip them here, with reasons. Link supporting sources near the claims they support, and distinguish sourced facts from your inferences and recommendations. If the evidence cannot support a recommendation, say what remains unknown.
Do not change code. The output feeds back into the discussion, not a to-do list.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Create, update, or review Architecture Decision Records (ADRs) using the project's conventions. Use when asked to record an architecture decision or work on an ADR; not for general design discussion without a requested record.
The English summary is being prepared. Showing the original description.
Create, edit, or review AGENTS.md and CLAUDE.md repository instructions. Use for agent instruction setup and structure advice, not ordinary code changes that merely read those files.
The English summary is being prepared. Showing the original description.
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking or requests an in-depth interview with dig.
The English summary is being prepared. Showing the original description.
Discuss ideas and design choices before implementation, without editing files.
The English summary is being prepared. Showing the original description.
Clone 済みのこの環境リポジトリのルートで実行し、setup.yaml(唯一の真実)と dotfiles/、skills/ を基準に端末上の dotfiles、Claude/Codex skills、関連 CLI 設定を安全に同期・更新・整理する。Use when the user asks to set up, reconcile, migrate, update, clean, prune, or merge this repository's dotfiles or skills with an existing terminal, especially when existing local files must be inspected before applying repository versions.
The English summary is being prepared. Showing the original description.
remote の rebase や force-push 後に diverge した gh-stack を診断し、local state を退避して remote に合わせて再構築する。履歴書き換えが疑われる non-fast-forward の調査と復旧に使う。通常の sync や一般の Git branch 管理には使わない。
The English summary is being prepared. Showing the original description.