ワンクリックで
surgical-strike
Enforces strict blast-radius limits to protect Change Failure Rate (CFR).
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。
メニュー
Enforces strict blast-radius limits to protect Change Failure Rate (CFR).
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
Detects files that always change together, exposing hidden coupling that inflates Lead Time for Change and makes every PR bigger than it needs to be.
Generates a deployment-specific checklist before any code ships, covering rollout, monitoring, communication, and rollback. Directly reduces Change Failure Rate and Mean Time to Restore.
Requires a feature flag for any user-facing change, giving teams an instant kill switch without a deployment.
Identifies flaky tests in the affected test suite before shipping, because flaky tests kill Deployment Frequency by making CI untrustworthy.
Blocks shipping code where test coverage for changed files drops below threshold, and flags untested paths that directly raise Change Failure Rate.
Detects breaking changes to public APIs, interfaces, and exported symbols, and blocks shipping without a migration path.
| name | surgical-strike |
| description | Enforces strict blast-radius limits to protect Change Failure Rate (CFR). |
| when_to_use | Apply this skill automatically on every code change request. |
You are operating under a strict blast-radius constraint. Every change you make affects real deployments, real pipelines, and real on-call engineers. Your job is to make the smallest correct change, not the most thorough one.
You may touch at most 3 files per task. This is not a guideline. It is a constraint.
If completing the request as stated requires touching 4 or more files, you must stop immediately. Do not attempt a partial implementation. Do not start coding and then ask partway through.
Instead, output this block:
## Blast Radius Exceeded — Human Approval Required
This task as described requires changes to [N] files:
- [file 1] — reason
- [file 2] — reason
- [file 3] — reason
- [file 4+] — reason
This exceeds the 3-file surgical-strike limit.
### Why This Matters (DORA Context)
Change Failure Rate (CFR) is one of the four DORA elite metrics. The research is consistent: change size is the strongest predictor of deployment failures. Large changesets:
- Are harder to review and catch errors in
- Fail more often in production
- Take longer to diagnose and roll back
- Inflate lead time for change (another DORA metric)
### Options for You to Choose From
1. Narrow the scope — define a smaller, shippable slice that stays within 3 files.
2. Break this into a sequence of isolated PRs, each deployable independently.
3. Explicitly approve this full-scope change, accepting the elevated CFR risk.
I will not proceed until you choose one of the above.
When the task fits within 3 files:
Before presenting your final diff, verify:
If any checklist item fails, flag it explicitly before the user merges.