
粘贴你的暂存改动、diff,或已合并的commit/PR列表,即可获得规范的Conventional Commits提交信息、按Keep a Changelog格式分组的更新日志条目,以及带理由的语义化版本(Semantic Versioning)升级建议(major/minor/patch)。专为使用Claude Code、Codex等编码智能体完成日常Git工作与发布流程的开发者打造。仅根据你粘贴的内容生成结果,无法判断的地方会标注[FILL IN]。无需API密钥,不访问仓库,不涉及个人数据。

粘贴一段git diff,即可获得一份可直接提交评审的Pull Request材料包:结构化的PR描述(概要、改动与原因、测试说明)、风险提示(数据库迁移、API/契约变更、配置改动、删除操作)、把最高风险的改动优先呈现给评审者的"评审关注地图",以及针对该diff量身定制的自查清单。专为使用Claude Code、Codex等编码智能体开发并提交PR的开发者打造。仅根据你粘贴的diff生成内容,无法判断的地方标注[FILL IN]。无需API密钥,不访问仓库,不涉及个人数据。

Paste a git diff and get a PR description, a changelog entry, and reviewer-risk callouts that name the file, what could break, and what to check. Only flags backed by the diff, and "no elevated-risk patterns found" when there are none. Built for AI-sized PRs.

Answer 5 questions. Get a finished CLAUDE.md or AGENTS.md you can drop into your repo. The questions: your exact build command, how tests run and what makes them pass silently, which paths must never be hand-edited, what your agent has gotten wrong before, and what looks wrong but is deliberate. Skip one and it becomes a TODO in the file, never a guess. Already have a file? Paste it and get it back without the lines your agent could read from the repo. Free trial: 3 requests in 1 day.

比較前後版本清單與依賴關係,找出變更影響及需要重審的項目。Compare declared revision dependencies and identify reviews needing renewal. Local Python download; no model required.

From a local ISSUE file to a verified, committed deliverable — with a fresh-subagent acceptance gate that catches wrong screenshots and leaked logic.

I've managed thousands of pull requests across open-source and enterprise repos — branching strategies, CI debugging, merge conflicts, and review workflows. This skill encodes that entire pipeline so you can automate your PR lifecycle through natural language, just tell me what you need.

交件前,找出漏檔、錯版與審核回條不一致。需要檔案與已接受的清單;Python 本機執行,不需模型。Catch missing files, wrong versions and mismatched review receipts. Local Python download; accepted manifest required. No model needed.

只读检查本地ZIP成员路径、别名、链接元数据、资源预算及CRC/SHA完整性,交付结构化清单。需自行提供CPython 3.11.1;不解包执行归档,不作恶意软件或来源认证结论。