debug
無料Investigate and fix bugs, failing tests, regressions, build failures, or unexpected behavior using boundary localization, evidence, and bounded hypotheses.
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。
Evaluate and act on code-review feedback with technical judgment. Use before applying external suggestions, especially when feedback is ambiguous, broad, or may conflict with repository constraints, and carry confirmed corrections to relevant consumers without widening scope.
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。
Treat review feedback as technical input to verify, not commands to obey blindly or social cues to praise.
Identify the requested change, claimed problem, affected behavior, urgency, and dependencies. Keep a single obvious correction inline; group related items by the behavior they change. Use the existing task record only when several dispositions need to remain reachable.
Do not implement a vague bundle such as “clean this up” without identifying the concrete behavior or quality concern.
For each item:
Classify the item:
Ask for clarification only when the missing answer materially changes behavior or scope and cannot be inferred safely.
Otherwise state the interpretation, choose the safest reversible implementation, and proceed.
A report may name one visible occurrence of a shared defect. After validating the concern:
Handle:
Test each meaningful behavior change. Batch tightly related items when separate changes would create temporary inconsistency.
Push back with evidence when:
State the technical reason and, when possible, offer a narrower alternative.
Avoid performative agreement. Useful responses include:
Report each item's disposition and the evidence or change associated with it. Do not claim all feedback is resolved when some items remain unverified or intentionally rejected.
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
Investigate and fix bugs, failing tests, regressions, build failures, or unexpected behavior using boundary localization, evidence, and bounded hypotheses.
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。
Choose responsibility for substantial work before deep execution: keep one responsibility's coupled parts in a single lane, and hand a bounded or independent lane to a worker when clean context, an unknown cause, independent review, or coordinator attention materially improves the outcome.
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。
Turn a rough feature, product interaction, or architecture idea into a concrete design. Use when meaningful decisions remain open; avoid for already-clear local edits.
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。
Execute an existing implementation plan or settled multi-step request through connected slices with targeted checks and controlled plan drift.
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。
Finish a development change safely by inspecting the final tree, verifying at the right scope, and performing only the requested Git or PR actions.
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。
Create an implementation plan sized to settled work. Use when a multi-step change benefits from sequencing, file targets, and explicit verification.
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。