scan-code-review-rules（扫码审核规则）

[IMPORTANT] Use TaskCreate to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.

Prerequisites: MUST ATTENTION READ before executing:

Scan & Update Reference Doc — Surgical updates only, never full rewrite. 1. Read existing doc first — understand current structure and manual annotations 2. Detect mode: Placeholder (only headings, no content) → Init mode. Has content → Sync mode. 3. Scan codebase for current state (grep/glob for patterns, counts, file paths) 4. Diff findings vs doc content — identify stale sections only 5. Update ONLY sections where code diverged from doc. Preserve manual annotations. 6. Update metadata (date, counts, version) in frontmatter or header 7. NEVER rewrite entire doc. NEVER remove sections without evidence they're obsolete.

Output Quality — Token efficiency without sacrificing quality. 1. No inventories/counts — AI can grep | wc -l. Counts go stale instantly 2. No directory trees — AI can glob/ls. Use 1-line path conventions 3. No TOCs — AI reads linearly. TOC wastes tokens 4. No examples that repeat what rules say — one example only if non-obvious 5. Lead with answer, not reasoning. Skip filler words and preamble 6. Sacrifice grammar for concision in reports 7. Unresolved questions at end, if any

Quick Summary

Goal: Scan project codebase for established conventions, lint rules, common patterns, and anti-patterns, then populate docs/project-reference/code-review-rules.md with actionable review rules and checklists. (content auto-injected by hook — check for [Injected:...] header before reading)

Workflow:

1. Read — Load current target doc, detect init vs sync mode
2. Scan — Discover conventions and patterns via parallel sub-agents
3. Report — Write findings to external report file
4. Generate — Build/update reference doc from report
5. Verify — Validate rules reference real code patterns

Key Rules:

• Generic — works with any language/framework combination
• Derive rules from ACTUAL codebase patterns, not generic best practices
• Every rule should have a "DO" example from the project and a "DON'T" counterexample
• Focus on project-specific conventions that differ from framework defaults

Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).

Scan Code Review Rules

Phase 0: Read & Assess

1. Read docs/project-reference/code-review-rules.md
2. Detect mode: init (placeholder) or sync (populated)
3. If sync: extract existing sections and note what's already well-documented

Phase 1: Plan Scan Strategy

Discover code quality infrastructure:

• Linter configs (.eslintrc, .editorconfig, stylecop.json, .prettierrc, ruff.toml)
• CI quality gates (build scripts, test requirements, coverage thresholds)
• Code analysis configs (SonarQube, CodeClimate, custom analyzers)
• Existing code standards docs (CONTRIBUTING.md, CODING_STANDARDS.md)
• Git hooks (pre-commit, husky configs)

Use docs/project-config.json if available for architecture rules and naming conventions.

Phase 2: Execute Scan (Parallel Sub-Agents)

Launch 3 Explore agents in parallel:

Agent 1: Backend Rules

• Grep for naming conventions (class suffixes, method prefixes, interface naming)
• Find common base classes and when they're used vs not used
• Discover error handling patterns (try-catch, Result types, error middleware)
• Find dependency injection patterns (registration conventions, lifetime choices)
• Look for anti-patterns (direct DB access from controllers, business logic in wrong layer)
• Identify logging conventions (structured logging, log levels, correlation IDs)

Agent 2: Frontend Rules

• Grep for component conventions (naming, file organization, template patterns)
• Find state management rules (what goes in store vs component vs service)
• Discover styling conventions (BEM, CSS modules, utility classes, naming)
• Find subscription/memory management patterns (cleanup, unsubscribe)
• Look for accessibility patterns (ARIA, semantic HTML, keyboard navigation)
• Identify performance patterns (lazy loading, change detection, memoization)

Agent 3: Architecture Rules

• Find layer boundaries (what imports what, dependency direction)
• Discover cross-service communication patterns (direct calls vs messages)
• Find shared code conventions (what's shared vs duplicated)
• Look for testing conventions (test naming, test organization, mock patterns)
• Identify security patterns (auth checks, input validation, output encoding)
• Find configuration patterns (env vars, config files, secrets management)

Write all findings to: plans/reports/scan-code-review-rules-{YYMMDD}-{HHMM}-report.md

Phase 3: Analyze & Generate

Read the report. Build these sections:

Target Sections

Content Rules

• Every rule must have a "DO" code example from the actual project
• Every rule should have a "DON'T" counterexample (real or realistic)
• Use file:line references for all code examples
• Prioritize rules by impact (bugs prevented, not style preferences)
• Decision trees can use markdown flowchart format or nested bullet lists

Phase 4: Write & Verify

1. Write updated doc with <!-- Last scanned: YYYY-MM-DD --> at top
2. Verify: 5 code example file paths exist (Glob check)
3. Verify: anti-pattern examples are realistic (not fabricated)
4. Report: sections updated, rules count, anti-patterns discovered

Closing Reminders

• IMPORTANT MUST ATTENTION break work into small todo tasks using TaskCreate BEFORE starting
• IMPORTANT MUST ATTENTION search codebase for 3+ similar patterns before creating new code
• IMPORTANT MUST ATTENTION cite file:line evidence for every claim (confidence >80% to act)
• IMPORTANT MUST ATTENTION add a final review todo task to verify work quality
• IMPORTANT MUST ATTENTION execute two review rounds (Round 1: understand, Round 2: catch missed issues)
• IMPORTANT MUST ATTENTION read existing doc first, scan codebase, diff, surgical update only. Never rewrite entire doc.
• IMPORTANT MUST ATTENTION follow output quality rules: no counts/trees/TOCs, rules > descriptions, 1 example per pattern, primacy-recency anchoring.