Claude-skill-registry codex-test-integration

Validate integration impact and regression risks via claude-delegator (Code Reviewer expert). Use for complex tasks or API integration.

install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/codex-test-integration" ~/.claude/skills/majiayu000-claude-skill-registry-codex-test-integration && rm -rf "$T"
manifest: skills/data/codex-test-integration/SKILL.md
source content

Codex Integration Validation (via claude-delegator)

When to use

  • complexity
    :
    complex
    (always)
  • OR
    apiSpecConfirmed == true && hasMockImplementation == true
  • Integration with external APIs
  • Multi-component changes

Procedure

  1. Summarize change scope, endpoints, and integration points
  2. Capture the context.md path (default:
    {tasksRoot}/{feature-name}/context.md
    ) and read relevant code
  3. Build delegation prompt using the 7-section format below (integration-focused)
  4. Try Codex first:
    • Call
      mcp__codex__codex
      (include Integration Reviewer instructions in developer-instructions)
    • If successful, proceed to step 6
  5. Fallback to Claude (if Codex unavailable):
    • Error conditions: "quota exceeded", "rate limit", "API error", "unavailable"
    • Claude directly performs integration review following the guidelines below
    • Add note:
      "codex-fallback: Claude performed integration review directly"
  6. Record regression risks and additional test scenarios
  7. If a saved report is needed, store the full review in
    {tasksRoot}/{feature-name}/archives/
    and keep only a short summary in
    context.md

Delegation Format

Use the 7-section format with integration focus:

TASK: Validate integration changes at [context.md path] for regression risks and contract compliance.

EXPECTED OUTCOME: Regression risk assessment with additional test scenarios.

CONTEXT:
- Integration to validate: [feature/API description]
- Changed files: [list of modified files]
- API endpoints affected:
  * [Endpoint 1: method, path, purpose]
  * [Endpoint 2: method, path, purpose]
- Integration points: [external systems, services, databases]

CONSTRAINTS:
- Must maintain backward compatibility
- Existing contracts must not break
- Technical stack: [languages, frameworks, API versions]

MUST DO:
- Identify regression risks across all integration points
- Verify contract compliance (request/response schemas)
- Check edge cases and error handling for each endpoint
- Assess performance implications of integration changes
- Identify missing test scenarios
- Check for proper error handling and retry logic

MUST NOT DO:
- Approve without checking all integration points
- Ignore backward compatibility concerns
- Skip edge case analysis

OUTPUT FORMAT:
Summary → Regression risks → Contract compliance → Edge cases → Performance concerns → Missing test scenarios → Additional tests needed

Tool Call

mcp__codex__codex({
  prompt: "[7-section delegation prompt with full context]",
  "developer-instructions": "[contents of code-reviewer.md]",
  sandbox: "read-only",  // Advisory mode - review only
  cwd: "[current working directory]"
})

For Implementation Mode (Add Tests)

If you want the expert to implement missing tests:

mcp__codex__codex({
  prompt: "[same 7-section format, but add: 'Implement the missing test scenarios identified']",
  "developer-instructions": "[contents of code-reviewer.md]",
  sandbox: "workspace-write",  // Implementation mode - can add test files
  cwd: "[current working directory]"
})

Output (patch)

notes:
  - "codex-integration: [PASS/FAIL], regression-risks=[count], extra-tests=[count]"