Add spec references to your PR description to enable traceability:

Spec Requirement by PR Type:

See PR template for full guidance.

This PR was reviewed by six AI agents in parallel, analyzing different aspects of the changes:

• Security Agent: Scans for vulnerabilities, secrets exposure, and security anti-patterns
• QA Agent: Evaluates test coverage, error handling, and code quality
• Analyst Agent: Assesses code quality, impact analysis, and maintainability
• Architect Agent: Reviews design patterns, system boundaries, and architectural concerns
• DevOps Agent: Evaluates CI/CD, build pipelines, and infrastructure changes
• Roadmap Agent: Assesses strategic alignment, feature scope, and user value

Now I have enough context. Let me analyze the PR.

Code Quality Analysis: Detect-CopilotFollowUpPR.ps1

Code Quality Score

Overall: 4/5

Impact Assessment

• Scope: Isolated (new script and test file, minor doc update)
• Risk Level: Low
• Affected Components: .claude/skills/github/ directory only

Findings

Recommendations

1. The $script: scoping pattern is valid but differs from peer scripts like Get-PRContext.ps1 which use local variables. Consider aligning for consistency.
2. The single jq filter improvement mentioned in the PR description (addressing Enhancement: Simplify jq pipeline in Get-CopilotAnnouncement #238) is correctly implemented at line 78.
3. Test file correctly extracts functions via regex for unit testing without needing to mock the entire script.

Verdict

VERDICT: PASS
MESSAGE: Well-structured recovery of lost functionality. Follows existing patterns for GitHub skill scripts. No architectural violations or blocking issues. Minor consistency observations do not block merge.

Now I have full context on the product roadmap and project goals. Let me analyze the PR strategically.

Strategic Alignment Assessment

Feature Completeness

• Scope Assessment: Right-sized
• Ship Ready: Yes
• MVP Complete: Yes
• Enhancement Opportunities: Integration with pr-comment-responder agent for automated triage of follow-up PRs

Impact Analysis

Concerns

Recommendations

1. Merge as-is. The code was previously developed and reviewed; recovery is lower risk than new development.
2. Consider adding this to the pr-comment-responder decision tree in a follow-up PR to automate duplicate PR handling.
3. The script addresses a real maintainer pain point (Copilot creating duplicate follow-up PRs) that affects repository hygiene.

Verdict

VERDICT: PASS
MESSAGE: Recovered feature fills GitHub skills gap. Code was previously developed with tests. Scope is appropriate for the problem. No strategic conflicts with roadmap priorities.

Now let me check if there are any existing workflow files that might run the Pester tests, and verify the script path references are correct:

Good, the script and module exist. Now let me validate the PowerShell script syntax and run the tests to verify they pass:

I'm getting permission denied errors, but I can still review the files statically. Let me complete my DevOps review based on what I've examined:

Pipeline Impact Assessment

CI/CD Quality Checks

Findings

Template Assessment

• PR Template: Adequate - PR follows established template
• Issue Templates: N/A - No changes
• Template Issues: None

Automation Opportunities

Recommendations

1. The new PowerShell script follows existing project patterns (uses GitHubHelpers.psm1, structured JSON output, proper error handling).
2. Tests are in the correct location (tests/) and follow Pester conventions.
3. The pester-tests.yml workflow already includes tests/** and .claude/skills/** in testable paths, so these new files will be tested automatically.

Verdict

VERDICT: PASS
MESSAGE: No CI/CD or build changes. New PowerShell skill script and tests follow existing patterns. Pester workflow will automatically run the new tests.

Now I have sufficient context. Let me complete my architectural review.

Design Quality Assessment

Overall Design Score: 4/5

Architectural Concerns

Breaking Change Assessment

• Breaking Changes: No
• Impact Scope: None (new script)
• Migration Required: No
• Migration Path: N/A

Technical Debt Analysis

• Debt Added: Low
• Debt Reduced: N/A (new feature)
• Net Impact: Neutral

ADR Assessment

• ADR Required: No
• Decisions Identified: None requiring ADR
• Existing ADR: Aligns with ADR-005 (PowerShell-only), ADR-017 (output schema consistency)
• Recommendation: N/A

The script follows established patterns. No new architectural decisions are introduced.

ADR Alignment Verification

Recommendations

1. Minor: Extract bot username to a module constant in GitHubHelpers for reuse across scripts
2. Minor: Test coverage looks comprehensive but integration testing would require API mocking

Verdict

VERDICT: PASS
MESSAGE: Design is sound. Follows ADR-005 (PowerShell-only) and ADR-017 (output schema). Clean separation of functions, proper module usage, appropriate exit codes. Minor style suggestions are non-blocking.

Security Review: PR feat/skills/recover-detect-copilot-followup

PR Type Classification

Findings

Analysis Summary

Reviewed Areas:

1. Shell Injection (CWE-78): [PASS] All gh CLI calls use parameters that are either integer-typed ($PRNumber) or derived from Resolve-RepoParams. The $Owner and $Repo values are validated through the GitHubHelpers module. No user-controlled strings are passed directly to shell execution.

2. Secret Detection: [PASS] No hardcoded credentials, API keys, or tokens detected. Script uses gh CLI which handles authentication via environment.

3. Input Validation: [PASS] $PRNumber is typed as [int], preventing injection via that parameter. String parameters $Owner and $Repo are resolved through module function.

4. Error Handling: [PASS] Uses 2>$null for stderr suppression and checks $LASTEXITCODE. Error messages do not leak sensitive data.

5. Dependencies: [PASS] Only dependency is GitHubHelpers.psm1 (internal module). No external packages added.

6. Test File: [PASS] Test file uses Invoke-Expression on extracted function content, but this is a controlled test pattern extracting from the known script path. No external input is evaluated.

Recommendations

None required. The code follows secure patterns:

• Integer typing prevents injection on the primary parameter
• Module-based parameter resolution for owner/repo
• Proper exit code checking
• No secrets or credentials in code

Verdict

VERDICT: PASS
MESSAGE: No security vulnerabilities detected. PowerShell script uses proper parameter typing, module-based validation, and secure gh CLI invocation patterns.

Based on my review of the PR files and test structure, I can now provide the QA assessment.

VERDICT: PASS
MESSAGE: CODE PR with 284-line PowerShell script has 141-line test file covering pattern matching, categorization logic, and syntax validation.

PR TYPE: CODE
FILES:

• CODE: .claude/skills/github/scripts/pr/Detect-CopilotFollowUpPR.ps1 (new, 284 lines)
• CODE: tests/Detect-CopilotFollowUpPR.Tests.ps1 (new, 141 lines)
• DOCS: .claude/skills/github/SKILL.md (documentation update)

EVIDENCE:

Test Coverage Assessment:

Quality Concerns:

Regression Risk Assessment:

• Risk Level: Low. New script with no breaking changes to existing functionality.
• Affected Components: .claude/skills/github/scripts/pr/ (new file only)
• Breaking Changes: None. Additive changes only.
• Required Testing: Pattern matching and categorization logic verified. Full integration requires authenticated gh CLI.

Test Quality Analysis:

• Tests verify behavior via assertions on return values
• Tests are isolated (no shared state between contexts)
• Test names are descriptive: "Matches valid Copilot follow-up branch pattern", "Empty diff string returns DUPLICATE category"
• Functions extracted from script content for isolated testing (clever pattern for testing internal functions)

Positive Observations:

1. Script follows existing module patterns (imports GitHubHelpers.psm1)
2. Uses Resolve-RepoParams for consistent repository inference
3. Structured JSON output matches project conventions
4. Documentation updated in SKILL.md with decision tree entry, script reference, and example

Walkthrough

This PR adds a new PowerShell script that detects Copilot-generated follow-up PRs, analyzes their relationship to original PRs, and recommends actions. The script includes helper functions for pattern detection, diff retrieval, and comparison logic plus comprehensive test coverage.

Changes

Sequence Diagram

sequenceDiagram
    actor User
    participant Script as Detect-CopilotFollowUpPR
    participant GH as GitHub API (gh CLI)
    participant Logic as Diff Analysis

    User->>Script: Invoke with PRNumber
    Script->>GH: Search for follow-up PRs<br/>(head ref: copilot/sub-pr-*)
    
    rect rgb(240, 248, 255)
    Note over Script,GH: If no follow-ups found
    GH-->>Script: Empty result
    Script-->>User: Return found=false
    end
    
    rect rgb(255, 248, 240)
    Note over Script,GH: If follow-ups found
    GH-->>Script: List of follow-up PRs
    
    loop For each follow-up PR
        Script->>GH: Fetch follow-up diff
        GH-->>Script: Unified diff
        Script->>GH: Fetch original PR commits
        GH-->>Script: Commit list
        Script->>Logic: Compare diffs
        Logic-->>Script: Similarity score +<br/>Category (DUPLICATE,<br/>LIKELY_DUPLICATE, etc.)
    end
    
    Script->>GH: Get Copilot announcement
    GH-->>Script: Announcement data
    
    Script->>Logic: Map category to<br/>recommendation
    Logic-->>Script: Recommendation<br/>(CLOSE, REVIEW,<br/>MERGE, MANUAL)
    
    Script-->>User: Return JSON with<br/>analysis + recommendation
    end

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Suggested reviewers

• rjmurillo

Walkthrough

Adds a new PowerShell script for detecting and analyzing Copilot follow-up PRs with helper functions for pattern matching, API calls, diff comparison, and categorization logic. Includes documentation update and comprehensive test coverage.

Changes

Sequence Diagram

sequenceDiagram
    participant User
    participant Script as Detect-CopilotFollowUpPR
    participant GitHubAPI as GitHub API
    participant Analysis as Analysis Engine

    User->>Script: Invoke with PRNumber
    Script->>GitHubAPI: Search for PRs with head:copilot/sub-pr-{PRNumber}
    GitHubAPI-->>Script: Return follow-up PRs list
    
    Script->>GitHubAPI: Fetch Copilot announcement comment
    GitHubAPI-->>Script: Return announcement details
    
    loop For each follow-up PR
        Script->>GitHubAPI: Fetch unified diff
        GitHubAPI-->>Script: Return diff content
        
        Script->>GitHubAPI: Fetch original PR commits
        GitHubAPI-->>Script: Return commit list
        
        Script->>Analysis: Compare diff vs commits
        Analysis-->>Script: Return category<br/>(DUPLICATE, LIKELY_DUPLICATE, etc.)
    end
    
    Script->>Script: Map categories to recommendations
    Script-->>User: Output JSON with results<br/>and global recommendation

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~45 minutes

Suggested reviewers

• rjmurillo