Token导航 LogoToken导航TokenDH.com
前端设计敏感数据github未标认证来源可访问许可证需确认审计提醒

pr-comments-address公关评论地址

Agent Skill

pr-comments-address 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

282

周安装

12

GitHub Stars

13

下载量

99
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:pr-comments-address(公关评论地址)
来源仓库:https://github.com/ravnhq/ai-toolkit
仓库路径:skills/pr-comments-address
安装命令:
npx skills add https://github.com/ravnhq/ai-toolkit --skill pr-comments-address
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/ravnhq/ai-toolkit --skill pr-comments-address

简介

pr-comments-address 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装,需结合原始 README 确认具体用法。
  • 安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

PR Comments Address

Read open review comments from a GitHub PR, intelligently triage them by severity and type, apply code fixes, and draft professional replies. This skill ensures no feedback is missed and all changes are intentional.

Workflow

Phase 1: Detect PR and Fetch Comments

Identify the PR from arguments or auto-detect from current branch:

# If PR number/URL provided as argument, use it
# Otherwise detect from current branch
gh pr view --json number,url,title,body,author

If no PR is found, stop and ask the user for a PR number or URL.

Fetch all review comments from the PR:

gh pr view <PR> --json reviews,comments,reviewThreads

Parse the JSON output to extract:

  • Comment author and timestamp
  • File and line number
  • Comment body
  • Whether the comment thread is resolved
  • Whether the comment is a review or standalone comment

Read any project-specific rules from CLAUDE.md or .claude/rules/ if they exist (e.g., code style, testing requirements, review conventions). Apply those rules when triaging comments.

Phase 2: Triage Comments by Category

Categorize every unresolved comment into exactly one category:

CategoryDefinitionActionExample
Must-fixRequired change, unambiguous, blocking mergeApply code change"Add error handling for null case"
Should-fixRecommended improvement, clear intent, improves qualityApply code change after confirmation"Extract this to a separate function for readability"
Clarification neededThe intent is unclear or the solution depends on contextAsk user for decision"Why does this use setTimeout instead of Promise?"
Discussion / QuestionNo code change needed, just needs a replyDraft reply only"What's the performance impact of this change?"
Praise / AcknowledgementPositive feedback or agreementBrief acknowledgement"Nice implementation!"

Triage decision tree:

  1. Is the change clearly defined? → Must-fix or Should-fix
  2. Is there any ambiguity about the solution? → Clarification needed
  3. Does it require a code change? → Must-fix or Should-fix; if no: Discussion/Question
  4. Is it subjective feedback? → Discussion/Question

Output a triage table showing every comment, its category, and the required action.

Phase 3: Resolve Ambiguities

Present all "clarification needed" comments to the user with full context (file, line number, comment text). For each one, explicitly ask: "What should be changed?"

Do not proceed until the user has answered every ambiguous comment.

Phase 4: Present the Change Plan

Show the full plan for every code change:

For must-fix and should-fix comments:

  • File path and line numbers
  • Current code (1-2 lines context)
  • Proposed change
  • Rationale (from the comment or context)

For discussion/question comments:

  • The question or feedback
  • Proposed reply

Ask the user: "Should I proceed with these changes?"

Wait for explicit user confirmation before making any edits.

Phase 5: Apply Code Fixes

For each approved code change:

  1. Read the full file first — never patch blindly from the comment
  2. Make the minimal change that addresses the comment — no scope creep
  3. Keep edits independent — apply changes one at a time to avoid conflicts
  4. Do not refactor or reformat outside the scope of the comment
  5. Verify the change is applied by reading the relevant lines after the edit

Example workflow for a single change:

cat src/auth.ts          # read file
                         # make the change (using Edit tool)
git diff src/auth.ts     # then verify

Phase 6: Verify Changes

After all code changes are applied, verify:

  1. File syntax is valid (no broken imports, missing brackets, etc.)
  2. Changes align with the comment intent
  3. No unintended side effects (test imports, unused variables, etc.)

Use git diff to review all changes at once:

git diff

Phase 7: Draft Replies

For every comment (must-fix, should-fix, clarification, discussion, praise), draft a reply:

For applied changes (must-fix, should-fix):

  • Confirm what was changed, factually and briefly
  • Example: "Done — extracted into a separate function validateUser() in auth.ts:45-60."

For clarifications now resolved:

  • Brief explanation of the decision
  • Example: "Using setTimeout here to avoid race conditions with pending state updates."

For questions/discussion:

  • Answer directly without over-explaining
  • Keep it professional and concise
  • Example: "Good catch — I'll verify the performance impact in the staging environment."

For praise/acknowledgements:

  • Brief, genuine acknowledgement
  • Example: "Thanks for the review!"

Important: Do not use filler phrases like "Great point!", "Thanks for catching that!", "Absolutely!", "Looking forward to your feedback!" These add noise without substance.

Phase 8: Present Reply Drafts

Show all draft replies to the user for review. Ask: "Should I post these replies?"

Wait for user approval before posting anything.

Phase 9: Post Replies

Once approved, post each reply to its corresponding comment thread:

gh pr comment <PR> --in-reply-to <COMMENT_ID> --body "<reply text>"

Or use the GitHub API directly if the command-line interface differs:

gh api repos/{owner}/{repo}/pulls/comments/{comment_id}/replies \
  -f body="<reply text>"

Confirm which replies were posted successfully.

Important Principles

  • Never guess ambiguity resolution — always ask the user explicitly
  • Confirm changes factually — "Done — added null check at line 42" not "Fixed the issue"
  • One file at a time — apply changes sequentially to avoid merge conflicts
  • No scope creep — fix the comment, nothing more
  • Professional tone — direct, factual, no corporate pleasantries
  • Do not resolve threads — let the reviewer do that after they read your replies
  • Do not push or open PRs — only apply code and draft replies

Examples

Positive Trigger

User: "address the review comments and fix the code feedback on PR #42"

Expected behavior: Fetch comments from PR #42, triage into must-fix / should-fix / clarification / discussion, present triage table, wait for user input, apply fixes after confirmation, draft replies.


User: "fix the PR feedback on my current branch"

Expected behavior: Detect PR from current branch, fetch comments, triage, present plan, apply approved changes, draft replies.


User: "respond to the code review on https://github.com/org/repo/pull/99"

Expected behavior: Fetch comments from PR #99, triage, resolve any ambiguities with user input, apply fixes and draft replies after confirmation.

Non-Trigger

User: "review this pull request #123"

Expected behavior: Do not use pr-comments-address. The user wants to review a PR, not address existing feedback. Use a PR review workflow instead.


User: "create a pull request for my branch"

Expected behavior: Do not use pr-comments-address. The user wants to create a new PR, not address comments on an existing one.

Troubleshooting

Skill Does Not Trigger

  • Error: The skill is not selected when user asks to address PR comments.
  • Cause: Request wording does not match trigger phrases (used "review PR" or "commit changes" instead of "address review", "fix feedback", "handle comments").
  • Solution: Rephrase with explicit keywords: "address review comments", "fix PR feedback", "respond to review feedback" and retry.

No PR Found for Current Branch

  • Error: gh pr view returns no PR for the current branch.
  • Cause: Current branch does not have an open PR, or branch has not been pushed.
  • Solution: Provide PR number or URL explicitly, or push the branch and create a PR first using agent-pr-creator.

No Unresolved Comments Found

  • Error: The PR has no open review comments.
  • Cause: All comments have already been resolved, or the PR has no reviews.
  • Solution: Inform the user that there are no pending comments. If they expected comments, verify they specified the correct PR.

GitHub CLI Not Authenticated

  • Error: gh commands fail with "not authenticated" or token errors.
  • Cause: GitHub CLI is not logged in or auth token has expired.
  • Solution: Run gh auth login to authenticate, then retry the skill.

Ambiguous Comment Intent

  • Error: Cannot determine what the reviewer wants changed.
  • Cause: Comment is vague, subjective, or depends on context not visible in the comment thread.
  • Solution: Present the comment to the user with full context (file, line, code snippet) and ask explicitly what change they want to make. Do not assume intent.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

32.65%
按下载量换算32

Claude

27.54%
按下载量换算27

Cursor

20.37%
按下载量换算20

Gemini CLI

9.79%
按下载量换算10

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills