メニュー
Use for both preparing work for review and responding to review feedback. Keep review evidence local, concrete, and scoped to the real contract. Verify technically before agreeing or changing code.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when asked to work in a continuous loop until done or truly blocked. Orchestrates pickup, planning, TDD execution, implementation waves when needed, debugging side-chains, verification, and resume behavior.
Use before planning when a feature or change request is underspecified. Resolve high-impact ambiguity around scope, consumers, security, failure handling, and acceptance criteria before writing the implementation plan.
Use before implementation. Merge design discovery and plan writing into one DR-specific flow: identify the consumer, inspect the repo, debate boundaries with partner and combatant, and produce an execution-ready plan that hands off cleanly into DR-native tasks, TDD, and implementation waves.
Use for the last mile: rerun verification, inspect diffs, commit or push if requested, and perform local docker-compose checks when needed. This is the DR local-tools replacement for MCP-style finish/deploy carryovers.
Use after planning and before implementation when you need a concrete, dependency-ordered task list. Turn the DR feature spec and plan into independently testable task slices with exact file paths and verification commands.
Use when resuming work in this repo after a pause or starting a fresh session on in-flight changes. Reconstruct the current state, open risks, relevant files, completed verification, and next actions.
| name | dr-code-review |
| version | 2026-03-31.1 |
| description | Use for both preparing work for review and responding to review feedback. Keep review evidence local, concrete, and scoped to the real contract. Verify technically before agreeing or changing code. |
Include:
For non-trivial reviews, also include:
Observed: concrete findings from code or verificationInferred: risk or behavior conclusionProposed: requested change or rebuttalConfidence: 0.0-1.0 with one-line reasonFalsifier: what would prove the review conclusion wrongReflection: weakest assumption and the next checkHelpful commands:
git diff --stat
git diff
Portal:
cd portal && go test ./...
cd portal && golangci-lint run ./...
Agent:
cd dr-agent && GOOS=windows go test ./...
cd dr-agent && GOOS=windows golangci-lint run ./...
Do this in order:
When responding to non-trivial comments, label your reasoning with Observed, Inferred, and Proposed.
Do not agree with review feedback before verification. The required pattern is:
For larger review items, prefer one issue at a time:
GOOS=windows constraintsFor non-trivial review disputes:
Use the canonical role behavior from docs/AGENT_ROLES.md.
For larger reviews, use narrow read-only subagents:
reviewersecurity_reviewer for security-sensitive changesUse the model and reasoning defaults from docs/CODEX_AGENT_POLICY.md rather than treating all review work as equal-cost and equal-depth.
For larger reviews, use narrow read-only subagents:
reviewer for correctness, regressions, and missing testssecurity_reviewer for security-sensitive diffswindows_agent_reviewer for dr-agent/ changesportal_explorer when the portal execution path is unclearDo not use subagents for tiny, obvious review fixes.