Processes code review feedback by classifying each finding, reproducing or validating it against the real contract, applying the smallest in-scope fix, and re-running focused proof. Use after comments arrive on a diff, pull request, design, or implementation. Not for requesting a first review or debugging an unrelated failure.
インストール方法を見る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 feedback as evidence to investigate, not as an attack to win or a list to obey blindly.
For a defect, reproduce the failure or inspect the real code path before editing. For a disputed finding, state the invariant and the evidence that supports the decision. Do not dismiss a comment because the suite is green.
Do not bundle unrelated cleanup or change the product contract to silence a reviewer. Use finding-log.md for multiple comments.
Every in-scope finding is fixed, verified, or explicitly accepted with a reason; out-of-scope work is recorded for follow-up; and the re-review boundary is clear.
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
Designs and runs reproducible evaluations for AI agents, prompts, tools, skills, and model-backed workflows using realistic datasets, isolated baselines, objective assertions, rubric grading, trajectory analysis, cost/latency tracking, and regression comparison. Use when measuring agent quality, optimizing skill triggering, comparing prompts or models, or gating an AI feature release. Not for ordinary deterministic unit tests.
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。
Designs or reviews HTTP, REST, GraphQL, RPC, CLI, webhook, event, and service interfaces with explicit inputs, outputs, errors, compatibility, idempotency, pagination, authentication, versioning, and observability. Use when introducing or changing an API or cross-component contract. Not for internal implementation details with no boundary or for debugging one API failure; use root-cause-debugging there.
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。
Reviews an existing codebase for structural friction, unclear ownership, leaky or shallow interfaces, excessive coupling, misplaced state, poor testability, and risky dependency direction, then prioritizes evidence-backed improvement candidates. Use for architecture audits, modularization, modernization, or recurring cross-cutting change pain. Not for designing one new interface, simplifying a local function, or fixing a reproduced bug.
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。
Validates a running application, CLI, API, service, or generated artifact as a user or operator against a prewritten observable behavior contract while remaining source-blind. Use for acceptance checks, runtime proof, anti-fake probes, release smoke tests, or an independent companion to code review. Not for source-quality findings, root-cause diagnosis, or visual design judgment outside the contract.
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。
Closes a completed development branch by checking the final diff and proof, presenting merge, pull-request, keep, or discard options, and cleaning up only after the user or repository workflow chooses a path. Use when feature work is complete and the branch must be integrated or retired. Not for claiming a feature is complete before 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。
Tests a real web user flow with a browser by asserting semantic behavior, network and loading states, keyboard access, responsive layouts, and stable visual evidence. Use for browser bugs, end-to-end UI behavior, responsive or accessibility checks, and screenshot baselines. Not for static source review without a browser or for backend-only tests.
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。