Agent skill
workflow-integration-ci
PR review response workflow - fetch comments, triage, and respond to review feedback (GitHub and GitLab)
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/workflow-integration-ci
SKILL.md
PR Workflow Skill (Provider-Agnostic)
Handles PR review comment workflows - fetching comments, triaging them, and generating appropriate responses. Works with both GitHub and GitLab via the unified tools-integration-ci abstraction.
What This Skill Provides
Workflows (Absorbs 2 Agents)
-
Fetch Comments Workflow - Retrieves PR review comments
- Uses
tools-integration-ciabstraction (GitHub or GitLab) - Replaces: review-comment-fetcher agent
- Uses
-
Handle Review Workflow - Processes and responds to comments
- Triages each comment for appropriate action
- Implements code changes or generates explanations
- Replaces: review-comment-triager agent
When to Activate This Skill
- Responding to PR review comments
- Processing review feedback
- Implementing reviewer-requested changes
- Generating explanations for reviewers
Workflows
Workflow 1: Fetch Comments
Purpose: Fetch all review comments for a PR.
Input: PR number (optional, defaults to current branch's PR)
Steps:
-
Get PR Comments via CI Integration
Use the
pr-commentscommand from marshal.json (provider-agnostic):bash# Resolve command from config COMMAND=$(jq -r '.ci.commands["pr-comments"]' .plan/marshal.json) eval "$COMMAND --pr-number {number} [--unresolved-only]"Or use the workflow script for additional processing:
bashpython3 .plan/execute-script.py pm-workflow:workflow-integration-ci:pr fetch-comments [--pr {number}]Output (TOON format):
toonstatus: success operation: pr_comments provider: github|gitlab pr_number: 123 total: N unresolved: N comments[N]{id,author,body,path,line,resolved,created_at}: c1 alice Fix security issue src/Auth.java 42 false 2025-01-15T10:30:00Z -
Return Comment List
Output: Structured list of comments for triage
Workflow 2: Handle Review
Purpose: Process review comments and respond appropriately.
Input: PR number or comment list from Fetch workflow
Steps:
-
Get Comments If not provided, use Fetch Comments workflow first.
-
Triage Each Comment For each unresolved comment:
Script:
pm-workflow:workflow-integration-cibashpython3 .plan/execute-script.py pm-workflow:workflow-integration-ci:pr triage --comment '{json}'Script outputs decision:
json{ "comment_id": "...", "action": "code_change|explain|ignore", "reason": "...", "priority": "high|medium|low|none", "suggested_implementation": "..." } -
Process by Action Type
For code_change:
- Read file at comment location
- Implement suggested change using Edit tool
- Reply to comment with commit reference
For explain:
- Generate explanation based on code context
- Reply to comment using gh CLI:
bash
gh pr comment {pr} --body "..."
For ignore:
- No action required
- Log as skipped
-
Group by Priority
- Process high priority first
- Then medium, then low
-
Return Summary
Output:
{
"pr_number": 123,
"processed": {
"code_changes": 3,
"explanations": 1,
"ignored": 1
},
"files_modified": ["..."],
"status": "success"
}
Scripts
Script: pm-workflow:workflow-integration-ci → pr.py
pr.py fetch-comments
Purpose: Fetch PR review comments from GitHub.
Usage:
python3 .plan/execute-script.py pm-workflow:workflow-integration-ci:pr fetch-comments [--pr <number>]
Requirements: gh CLI installed and authenticated
Output: JSON with comments array
pr.py triage
Purpose: Analyze a single comment and determine action.
Usage:
python3 .plan/execute-script.py pm-workflow:workflow-integration-ci:pr triage --comment '{"id":"...", "body":"...", ...}'
Output: JSON with action decision
References (Load On-Demand)
Review Response Guide
Read references/review-response-guide.md
Provides:
- Comment classification patterns
- Response templates
- Best practices for reviewer communication
Comment Classification
| Pattern | Action | Priority |
|---|---|---|
| security, vulnerability, injection | code_change | high |
| bug, error, fix, broken | code_change | high |
| please add/remove/change | code_change | medium |
| rename, variable name, typo | code_change | low |
| why, explain, reasoning, ? | explain | low |
| lgtm, approved, looks good | ignore | none |
Integration
Commands Using This Skill
- /pr-handle-pull-request - Full PR workflow
- /pr-respond-to-review-comments - Comment response
Related Skills
- sonar-workflow - Often used together in PR workflows
- git-workflow - Commits changes after responses
Quality Verification
- Self-contained with relative path pattern
- Progressive disclosure (references loaded on-demand)
- Scripts output TOON/JSON for machine processing
- Both fetcher and triager agents absorbed
- Clear workflow definitions
- Provider-agnostic via tools-integration-ci
References
- tools-integration-ci:
plan-marshall:tools-integration-ciskill - GitHub CLI: https://cli.github.com/
- GitLab CLI: https://gitlab.com/gitlab-org/cli
- Code Review Best Practices: https://google.github.io/eng-practices/review/
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?