pr-comment-responder

rjmurillo 的头像作者 rjmurillo

PR审查协调员,收集评论上下文,认可每一份反馈,并系统性地处理所有审查者的评论。按可执行性进行分类,跟踪对话线程,并将每个评论映射到解决状态。在处理PR反馈、审查线程或机器人评论时使用。

📖 中文导读:PR审查协调员,收集评论上下文,认可每一份反馈,并系统性地处理所有审查者的评论。按可执行性进行分类,跟踪对话线程,并将每个评论映射到解决状态。在处理PR反馈、审查线程或机器人评论时使用。 以下正文为该技能的原始 SKILL.md 内容(英文),上方安装命令可直接使用。

name: pr-comment-responder version: 1.0.0 description: PR review coordinator who gathers comment context, acknowledges every piece of feedback, and ensures all reviewer comments are addressed systematically. Triages by actionability, tracks thread conversations, and maps each comment to resolution status. Use when handling PR feedback, review threads, or bot comments. license: MIT model: claude-sonnet-4-5 metadata: argument-hint: Specify the PR number or review comments to address

PR Comment Responder

Coordinates PR review responses through context gathering, comment tracking, and orchestrator delegation.

Triggers

Phrase Action
"respond to PR comments" Full workflow
"address review feedback" Full workflow
"handle PR #123 comments" Target specific PR
"Review PR 806 comments... github.com/.../pull/806" Extract PR from text/URL

Quick Reference

Context Inference (Phase -1)

ALWAYS extract PR context from prompt first. Never prompt for information already provided.

# Extract PR number and owner/repo from user prompt
$context = pwsh .claude/skills/github/scripts/utils/Extract-GitHubContext.ps1 -Text "[prompt]" -RequirePR

Supported patterns:

  • Text: PR 806, PR #806, pull request 123, #806
  • URLs: github.com/owner/repo/pull/123

See references/workflow.md Phase -1 for full details.

Tools

Operation Script
Context extraction Extract-GitHubContext.ps1
PR metadata Get-PRContext.ps1
Comments Get-PRReviewComments.ps1 -IncludeIssueComments
Reviewers Get-PRReviewers.ps1
Reply Post-PRCommentReply.ps1
Reaction Add-CommentReaction.ps1
Resolve thread Resolve-PRReviewThread.ps1

Reviewer Priority

Priority Reviewer Signal
P0 cursor[bot] 100% actionable
P1 Human reviewers High
P2 coderabbitai[bot] ~50%
P2 Copilot ~44%

Workflow Phases

  1. Context inference: Extract PR number from prompt (BLOCKING)
  2. Memory init: Load pr-comment-responder-skills memory
  3. Context gather: PR metadata, reviewers, all comments
  4. Acknowledge: Batch eyes reactions
  5. Generate map: .agents/pr-comments/PR-[N]/comments.md
  6. Delegate: Each comment to orchestrator
  7. Implement: Via orchestrator delegation
  8. Verify: All comments addressed, CI passing

See references/workflow.md for full phase details.

Verification Gates

Before completion, verify:

  • All comments resolved (COMPLETE or WONTFIX)
  • No new comments after 45s wait
  • CI checks passing
  • All threads resolved
  • Commits pushed

See references/gates.md for gate implementation.

Response Templates

See references/templates.md for:

  • Won't Fix responses
  • Clarification requests
  • Resolution replies

Bot Handling

See references/bots.md for:

  • Copilot follow-up PR handling
  • CodeRabbit commands
  • cursor[bot] patterns