Back to list
Z-M-Huang

review-codex

by Z-M-Huang

Claude (planner + coder) and codex (reviewer)

13🍴 4📅 Jan 24, 2026

SKILL.md


name: review-codex description: Final code/plan review using Codex. Use as the last review step after sonnet and opus. plugin-scoped: true context: fork allowed-tools: Read, Glob, Grep, Bash, Write

Review Codex (Final Review)

You are the final reviewer, invoking Codex for the ultimate review before approval.

Reference

Read ${CLAUDE_PLUGIN_ROOT}/docs/standards.md for the complete review checklist.

Your Role (Final Gate)

  • Final gate: Last check before plan approval or code completion
  • External review: Use Codex CLI for independent assessment
  • Comprehensive: Verify all previous review findings are addressed
  • OWASP verification: Final security gate for OWASP Top 10

Determine Review Type

Check which files exist:

  1. If .task/plan-refined.json exists and no .task/impl-result.jsonPlan Review
  2. If .task/impl-result.json exists → Code Review

Session Management

Check if .task/.codex-session-active exists:

  • If yes: This is a subsequent review (Codex has reviewed before)
  • If no: This is a first review

Codex Review Checklist

The Codex prompt should verify all aspects from standards.md:

Security - OWASP Top 10 (Final Gate)

  • All 10 OWASP categories verified
  • No injection vulnerabilities
  • Authentication/authorization properly implemented
  • No sensitive data exposure
  • No known CVEs in dependencies

Error Handling (Completeness)

  • All failure paths handled
  • No sensitive data in errors
  • Graceful degradation

Resource Management (Verification)

  • No memory/connection leaks
  • Proper cleanup in all paths

Configuration (Overall)

  • No hardcoded secrets
  • Environment-based configuration

Code Quality (Clarity)

  • Readable and maintainable
  • Appropriate complexity
  • Adequate documentation
  • No unnecessary duplication

Concurrency (Verification)

  • No race conditions
  • Thread safety verified

Logging (Completeness)

  • Critical operations logged
  • No secrets in logs

Dependencies (Final Check)

  • Security audit passed
  • No unnecessary deps

API Design (Overall)

  • Consistent responses
  • Proper validation

Backward Compatibility (Migration)

  • Breaking changes documented
  • Migration path clear

Testing (Coverage)

  • Adequate test coverage
  • Tests pass

Over-Engineering (Balance)

  • No unnecessary abstractions
  • Appropriate complexity for the problem
  • No premature optimization

For Plan Reviews

  1. Read .task/plan-refined.json
  2. Read ${CLAUDE_PLUGIN_ROOT}/docs/standards.md for review criteria
  3. Build prompt for Codex
  4. Invoke Codex using Bash:
codex exec \
  --full-auto \
  --output-schema "${CLAUDE_PLUGIN_ROOT}/docs/schemas/plan-review.schema.json" \
  -o .task/review-codex.json \
  "Review the plan in .task/plan-refined.json against ${CLAUDE_PLUGIN_ROOT}/docs/standards.md. As the final gate reviewer, verify: (1) OWASP Top 10 security considerations are addressed, (2) Error handling strategy is complete, (3) Resource management is considered, (4) No hardcoded secrets planned, (5) Code quality approach is sound, (6) Testing strategy is adequate, (7) No over-engineering - complexity is appropriate for the problem. Check for completeness, feasibility, and potential issues. If requirements are ambiguous or missing information that cannot be inferred, set needs_clarification: true and provide clarification_questions."

For Code Reviews

  1. Read .task/impl-result.json
  2. Read ${CLAUDE_PLUGIN_ROOT}/docs/standards.md for review criteria
  3. Build prompt for Codex
  4. Invoke Codex using Bash:
codex exec \
  --full-auto \
  --output-schema "${CLAUDE_PLUGIN_ROOT}/docs/schemas/review-result.schema.json" \
  -o .task/review-codex.json \
  "Review the implementation in .task/impl-result.json against ${CLAUDE_PLUGIN_ROOT}/docs/standards.md. As the final gate reviewer, verify ALL categories: (1) OWASP Top 10 - check all 10 security categories, (2) Error handling completeness, (3) Resource management - no leaks, (4) Configuration - no hardcoded secrets, (5) Code quality - readability, simplification, comments, DRY, (6) Concurrency safety if applicable, (7) Logging hygiene, (8) Dependency security, (9) API design consistency, (10) Backward compatibility, (11) Test coverage, (12) Over-engineering - no unnecessary complexity. If requirements are ambiguous or behavior is unclear and cannot be inferred from context, set needs_clarification: true and provide clarification_questions."

For Subsequent Reviews

If .task/.codex-session-active exists, use resume. Branch on review type:

Plan Re-Review (no impl-result.json)

codex exec \
  --full-auto \
  --output-schema "${CLAUDE_PLUGIN_ROOT}/docs/schemas/plan-review.schema.json" \
  -o .task/review-codex.json \
  resume --last \
  "Re-review the plan changes. Previous concerns should be addressed. Verify all review categories from standards.md: OWASP Top 10 security, error handling, resource management, configuration, code quality, testing strategy, and no over-engineering. If requirements are still ambiguous or missing information, set needs_clarification: true and provide clarification_questions."

Code Re-Review (impl-result.json exists)

codex exec \
  --full-auto \
  --output-schema "${CLAUDE_PLUGIN_ROOT}/docs/schemas/review-result.schema.json" \
  -o .task/review-codex.json \
  resume --last \
  "Re-review the code changes. Previous issues should be addressed. Verify all 12 review categories from standards.md: OWASP Top 10, error handling, resource management, configuration, code quality, concurrency, logging, dependencies, API design, backward compatibility, testing, and over-engineering. If requirements are still ambiguous or behavior unclear, set needs_clarification: true and provide clarification_questions."

After Codex Completes

  1. Mark session as active: touch .task/.codex-session-active
  2. Read .task/review-codex.json to get the result
  3. Check for needs_clarification: true - if set, the autonomous pipeline will pause for user input
  4. Report back:
    • Review type (plan or code)
    • Status from Codex (approved or needs_changes)
    • Summary of Codex findings
    • If needs_clarification, include the clarification_questions
    • Confirm output in .task/review-codex.json

If Codex Fails

If the Bash command fails or output is invalid:

  1. Report the error to the user
  2. Check if codex CLI is installed and authenticated
  3. Try the command again with verbose output
  4. If persistent failure, ask user to verify Codex CLI setup

Score

Total Score

65/100

Based on repository quality metrics

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

+10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

1ヶ月以内に更新

+10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

+5

Reviews

💬

Reviews coming soon