本文へ移動
cccskills
無料GitHub で公開

reviewer-correctness

Verifies bug fixes are correct and complete by tracing callers, checking imports/exports, validating path resolution, and confirming no regressions. Does the code work? Use when reviewing code changes for functional correctness.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md4.6 KB

SKILL.md(原文)

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。

Correctness Reviewer

Verifies that code changes are functionally correct by tracing callers, checking imports/exports, validating path resolution, and confirming no side effects or regressions.

Review Complete When

  • Every modified function has >=1 caller audit (traced, compatibility verified)
  • All imports/exports reconciled (no orphaned, no missing)
  • Circular dependencies validated or ruled out for all affected modules
  • Path resolution (if applicable) verified relative to correct base
  • No regression risks identified in unchanged functions
  • Summary table shows status for all checks
  • Overall verdict is clear ([PASS] safe to merge / [WARN] concerns listed / [FAIL] needs work)

A review is not complete if any check has an "unknown" or "not checked" status.

Instructions

1. Verify the Fix Itself

For each change, confirm the code modification is syntactically and semantically correct:

# Check the change is present
grep -n "export function createSession" agent-spawning.ts
grep -n "path.join(consolidationsDir, f)" shared.ts

Check:

  • Required imports are present (e.g., import * as path from "node:path")
  • The change uses the correct API (e.g., path.join not string concatenation)
  • No syntax errors introduced

2. Trace All Callers

For each modified function, find and verify all callers:

# Find all callers
grep -rn "functionName" --include="*.ts" --include="*.js" | grep -v "function functionName"

For each caller, verify:

  • The caller's usage is compatible with the change
  • The caller receives the expected data format (e.g., full paths, not bare filenames)
  • No caller is broken by the change

3. Verify Imports and Exports

For export changes, confirm the full import chain:

# Check the export exists
grep -n "export function.*" target.ts

# Check the import exists
grep -n "import.*from.*target" consumer.ts

Verify:

  • All functions that need to be exported are exported
  • All consumers import the correct names
  • No orphaned imports (importing something that doesn't exist)
  • No missing exports (something imported but not exported)

4. Check Path Resolution

For path-related changes, verify:

# Check path module is imported
grep "import.*path" file.ts

# Verify path.join usage
grep "path.join" file.ts

# Check if paths are absolute
# (may require running the code or checking the logic)

Verify:

  • path module is imported
  • path.join or equivalent is used (not string concatenation)
  • Paths are resolved relative to the correct base directory
  • All consumers expect the same path format

5. Check for Circular Dependencies

Build the import chain for affected modules:

# Check imports in each file
grep "^import" shared.ts agent-spawning.ts dream-execution.ts

Verify:

  • No circular import chains exist
  • Import direction is consistent (no back-references)
  • The change doesn't introduce new coupling

6. Verify No Regressions

Check that existing functionality is preserved:

# Check unchanged functions still work
grep -n "checkConsolidationFreshness" shared.ts

Verify:

  • Unchanged functions are unaffected
  • No function signatures changed unexpectedly
  • No side effects on unrelated modules
  • The change is minimal and targeted

7. Report Findings

Structure the review as:

## Fix N: [Description]

### Correct: [PASS]/[WARN]/[FAIL] [Fix is correct/incomplete/incorrect]

**Evidence:**
- [file:line] `code snippet` -- [explanation]
- [file:line] `code snippet` -- [explanation]

**Callers verified:**
| Caller | Location | Uses correct format? | Status |
|--------|----------|---------------------|--------|
| [caller] | [file:line] | [yes/no] | [PASS]/[FAIL] |

**Independent implementations (not affected):**
- [module]: [brief note on why unaffected]

---

## Summary

| Check | Result |
|-------|--------|
| Fix N: [description] | [PASS]/[WARN]/[FAIL] [brief note] |
| Other callers affected | [PASS]/[FAIL] [note] |
| Required imports present | [PASS]/[FAIL] [note] |
| Circular dependency risk | [PASS]/[FAIL] [note] |

**[Overall verdict]**

Use status symbols:

  • [PASS] -- Verified correct, no issues
  • [WARN] -- Partially correct, minor concern
  • [FAIL] -- Incorrect, needs fixing

レビュー

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。

同じリポジトリのスキル

概要と使いどころ

Create and maintain AGENTS.md files for project onboarding and grounding. Use when initialized in a new project, when project structure changes significantly, or when creating subfolder-level AGENTS.md files for complex monorepos.

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。

AndrewBrown-USDA/aigb-skills-usda52026年10月10日 更新

Audit and update an arbitrary codebase to follow the Code.mil federal open-source licensing model (MIT with 17 U.S.C. Section 105 disclaimers and INTENT.md), including CRAN/R package DESCRIPTION setup.

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。

AndrewBrown-USDA/aigb-skills-usda52026年10月10日 更新

caveman

無料

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

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。

AndrewBrown-USDA/aigb-skills-usda52026年10月10日 更新

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes - Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/spec asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".

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。

AndrewBrown-USDA/aigb-skills-usda52026年10月10日 更新

Reads Copilot conversation history and extracts executive-ready success stories. Use when the user asks to study prior Copilot sessions, summarize what was done before, mine evidence, or quantify time, cost, risk, and priority alignment from GitHub Copilot history in any workspace or repository.

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。

AndrewBrown-USDA/aigb-skills-usda52026年10月10日 更新

Comprehensive investigation of complex multi-domain narratives (technology, theology, policy, finance). Systematically decouples factual claims from speculation through iterative verification, multi-source comparison, and contradiction resolution. Use for investigations requiring institutional knowledge, historical context, and handling of conflicting incentives.

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。

AndrewBrown-USDA/aigb-skills-usda52026年10月10日 更新

AndrewBrown-USDA のスキルをすべて見る

このスキルの問題を報告する