Treat issues found in review, testing, or runtime as traceable work tickets. The input is one or more findings; the output is an item-by-item handling record, fixes verified by regression, and explicit reasons and follow-up paths for items that cannot be fixed.
Scope of Application
Applies to structured findings, failing tests, or locatable defects in ordinary engineering repositories. If the task is constrained by an organization's proprietary status, approval, or archival rules, first load the corresponding binding layer; this skill assumes no specific directory, project, or toolchain.
Even a single obvious small change should retain a minimal record: reproduction evidence, fix, relevant regression, and receipt.
Customer-Readable Description
What This Skill Can Do
| Customer Wants | Execute Closed Loop | Customer Can See |
|---|---|---|
| Fix review comments one by one | Aggregate, decide, fix, and verify item by item | Status and evidence for each finding |
| Confirm whether closure is possible | Independently re-check fixes and regression results | Conclusions of pass, needs modification, or blocked |
| Defer some issues | Provide traceable follow-up locations and reasons | Clear deferral boundaries |
How Customers Use It
Provide findings, the target workspace, relevant tests or reproduction information, and the allowed scope of changes. If an item may involve deletion, overwriting, remote state, release, or scope expansion, obtain explicit authorization first; when information is insufficient, first compile an input checklist and do not begin modifications.
Dependencies and Installation
This skill requires no dedicated installation or private configuration. Its hard dependencies are accessible findings, the target workspace, and verification entry points commensurate with the risk; follow the target repository's existing contribution instructions and inspection conventions.
Logs and Completion Receipt
The process records the number of inputs, item-by-item decisions, verification status, and blocking reasons. The final receipt must cover the conclusion for each finding, the checks actually run, unhandled items, and residual risks; credentials or local private information must not be output.
Five-Step Fix Loop
1. Reproduce and Normalize Input
Read all findings and extract unique identifiers, severity, location, problem description, and source. Merge duplicates; when the same issue has different severities, use the higher level while preserving the sources. Run a minimal reproduction for each key issue, or record why it cannot currently be reproduced.
| Finding | 严重性 | 来源 | 位置 | 复现证据 |
|---|---|---|---|---|
| F-01 | high | review-a | <file:line> | <命令或步骤> |
2. Decide Item by Item Before Fixing
Each finding must be assigned a status before implementation:
| Decision | Meaning | Required Basis |
|---|---|---|
| fix | Fix in this round | Target behavior and verification plan |
| reject | Finding does not hold | Testable counter-evidence |
| defer | Do not fix for now | Traceable follow-up location, owner, or trigger condition |
Issues must not be silently omitted; 'look at it later' must not replace defer. For high-risk decisions or fixes that expand scope, first explain the impact to the user and obtain the necessary authorization.
3. Implement the Minimal Fix
Modify only the parts needed to make the finding satisfy the target behavior. It is forbidden to fake a fix with TODO, comments, suppressing exceptions, weakening assertions, or deleting tests. Each fix should have corresponding evidence in the diff; if multiple fixes depend on each other, state the dependencies explicitly.
After modification, search for similar patterns and decide whether to include them in this round; if excluded, record the scope rationale to avoid misrepresenting unchecked similar issues as resolved.
4. Regression and independent review
For each fix item, run at least one direct verification and run regression checks proportionate to the change surface. For each reject item, review its rebuttal evidence; for each defer item, review that the follow-up chain genuinely exists and is sufficiently specific.
In addition to testing, attack the conclusion through independent paths, such as checking the diff, constructing counterexample inputs, verifying callers, or having someone not involved in the implementation review it. If regressions, an empty diff, TODO fake fixes, invalid rejection reasons, or broken follow-up chains appear, loop back to step 2 or step 3.
5. Close out and respond
Only when all items have fix、reject or defer evidence can it be closed out. Use one of the following conclusions:
| Conclusion | Condition |
|---|---|
| approved | Fix passes, rejection reasons hold, deferred chains are valid |
| changes requested | There are unverified fixes, omissions, or decisions that do not hold |
| blocked | Necessary permissions, environment, input, or follow-up ownership are missing |
修复回执:<approved / changes requested / blocked>。
处理清单:<每条 finding 的决策与结果>
验证:<复现、直接检查、回归和独立复核>
未处理项:<defer 或 blocked 的原因和承接路径>
风险:<仍未覆盖的情形;无则写“无”>
Anti-omission and fake-fix gates
- Every finding in the input list has an explicit decision;
- Every fix has diff evidence of code or configuration behavior;
- Direct verification and regression checks are actually run, and failures are not hidden;
- Rejections and deferrals both have reviewable grounds;
- Unrelated files or unrelated refactoring are not mixed into the fix commit.
Security boundaries
Preserve user confirmation for deletion, overwrite, remote writes, release, and expanding the fix scope. Do not put local absolute paths, accounts, credentials, or private project context into records or examples.