Token导航 LogoToken导航TokenDH.com
开发操作浏览器github未标认证来源可访问许可证需确认审计通过

release-notes发行说明

Agent Skill

release-notes 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

1,454

周安装

60

GitHub Stars

103

下载量

475
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:release-notes(发行说明)
来源仓库:https://github.com/borghei/claude-skills
仓库路径:skills/release-notes
安装命令:
npx skills add https://github.com/borghei/claude-skills --skill release-notes
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/borghei/claude-skills --skill release-notes

简介

release-notes 将技术变更转化为用户导向的发行说明,适配多场景沟通需求。

  • 支持产品发布、迭代演示与客户通知等多种输出格式与语气风格。
  • 基于原始数据自动生成可读性强、重点突出的版本摘要。
  • 生成内容需经人工校验以确保准确性与品牌一致性。
  • 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

Release Notes Expert

Overview

Transform raw technical changes -- tickets, changelogs, git logs, PRDs -- into clear, user-benefit-oriented release notes. This skill ensures every release communicates value to the right audience in the right tone.

When to Use

  • Product Releases -- Announcing new versions to customers, partners, or internal stakeholders.
  • Sprint Demos -- Summarizing what shipped for sprint review audiences.
  • Changelog Maintenance -- Keeping a running log of changes across releases.
  • Customer Communication -- Preparing release announcements for email, in-app, or documentation.

Methodology

Step 1: Gather Raw Input

Collect all changes from the release cycle:

  • Jira/Linear tickets -- Completed stories, bugs, and tasks
  • Git log -- Merge commits since last release tag
  • PRD references -- Feature specs that shipped
  • Hotfix records -- Emergency fixes deployed between releases

Step 2: Classify Each Change

Assign every change to exactly one category:

CategoryDefinitionExample
New FeaturesNet-new capabilities that did not exist beforeNew export-to-PDF option
ImprovementsEnhancements to existing functionalityFaster dashboard loading
Bug FixesCorrections to broken or incorrect behaviorFixed login redirect loop
Breaking ChangesChanges that require user action to adaptAPI v2 replaces v1 endpoints
DeprecationsFeatures scheduled for future removalLegacy CSV import will be removed in v4.0

Classification rules:

  • If a change adds something entirely new, it is a New Feature.
  • If it makes something existing better (faster, easier, more reliable), it is an Improvement.
  • If it fixes something that was wrong, it is a Bug Fix.
  • If users must change their behavior, configuration, or integration, it is a Breaking Change.
  • If a feature still works but will be removed later, it is a Deprecation.

Step 3: Rewrite for User Benefit

The most critical step. Every entry must lead with the benefit to the user, not the technical change.

Rewriting principles:

  1. Lead with the outcome. What can the user do now that they could not before, or what is better for them?
  2. Use plain language. Avoid internal jargon, code references, or implementation details.
  3. Keep it to 1-3 sentences. One sentence for minor items, up to three for significant features.
  4. Include context when needed. If users need to take action, tell them exactly what to do.

Before and after examples:

Technical (Bad)User-Benefit (Good)
Implemented Redis caching layer for dashboard queriesDashboards now load up to 3x faster
Refactored authentication module to use OAuth 2.0 PKCE flowSign-in is now more secure and works reliably across all browsers
Fixed null pointer exception in report export handlerReport exports no longer fail when date ranges include empty days
Migrated user preferences API from v1 to v2 schemaAction required: Update your API calls to use the new /v2/preferences endpoint by April 30. See migration guide.
Added feature flag for beta dashboardYou can now opt into the redesigned dashboard from Settings > Beta Features

Red flags that an entry needs rewriting:

  • Mentions a class name, function, or library
  • Starts with "Refactored," "Migrated," or "Updated" without stating impact
  • Uses acronyms the target audience would not know
  • Describes what the team did instead of what the user gains

Step 4: Adjust Tone for Audience

AudienceToneStyle Notes
B2B / EnterpriseProfessional, preciseEmphasize reliability, security, compliance. Avoid casual language.
ConsumerFriendly, conversationalUse "you" and "your." Celebrate new features. Keep it light.
Developer / APITechnical, directInclude endpoint names, SDK versions, code snippets. Be specific.
InternalDetailed, context-richInclude ticket IDs, team names, technical details as needed.

Step 5: Assemble the Release Notes

Use the output template below. Include only categories that have entries -- do not show empty sections.

Output Template

# [Product Name] v[X.Y.Z] Release Notes

**Release Date:** [YYYY-MM-DD]

---

## New Features

- **[Feature Name]** -- [1-3 sentence description of user benefit]. ([TICKET-ID])

## Improvements

- **[Improvement Name]** -- [1-2 sentence description of what is better]. ([TICKET-ID])

## Bug Fixes

- **[Bug Fix Name]** -- [1 sentence describing what was broken and that it is now fixed]. ([TICKET-ID])

## Breaking Changes

> **Action Required:** The following changes require updates on your end.

- **[Change Name]** -- [Description of what changed and exactly what the user must do]. ([TICKET-ID])

## Deprecations

> **Planned Removal:** The following features will be removed in a future release.

- **[Feature Name]** -- [What is being deprecated and when it will be removed. Recommend alternative if available]. ([TICKET-ID])

---

**Full changelog:** [link]
**Questions?** [support link or contact]

Python Tool

Use scripts/release_notes_generator.py to generate formatted release notes from structured input.

# Generate from JSON input
python scripts/release_notes_generator.py --input changes.json --product-name "Acme App" --version "2.5.0"

# Run with demo data
python scripts/release_notes_generator.py --demo --product-name "Acme App" --version "1.0.0"

# Output as JSON instead of markdown
python scripts/release_notes_generator.py --input changes.json --format json --product-name "Acme App" --version "2.5.0"

See scripts/release_notes_generator.py --help for full usage.

Integration with Other Skills

  • Use summarize-meeting/ to capture release planning discussions.
  • Use job-stories/ or wwas/ to trace features back to their original motivation.
  • Pair with ../senior-pm/ for stakeholder communication planning around major releases.

Troubleshooting

SymptomLikely CauseResolution
Tool flags too many entries as "technical language"TECHNICAL_PATTERNS regex is broad, catching common words like "update" or "add"Review the flagged entries; the tool provides suggestions, not mandates -- ignore false positives for user-facing descriptions
All entries classified as same typeInput JSON uses wrong type values or inconsistent casingUse exact lowercase types: feature, improvement, bugfix, breaking, deprecation
Empty sections appear in outputNo entries of that type exist but template still renders the sectionThe tool only renders sections with entries; empty sections indicate a data issue in input
Breaking changes not highlighted prominentlyEntries marked as improvement instead of breakingReview classification rules: if users must change behavior, config, or integration, it is breaking, not improvement
Release notes sound like commit messagesDescriptions written from developer perspective, not user perspectiveApply the rewriting principles: lead with outcome, use plain language, 1-3 sentences per entry
--demo flag requires --product-name and --versionThese are required arguments regardless of input sourceAlways provide both: --demo --product-name "App" --version "1.0.0"
JSON output missing rewriting suggestionsNo entries triggered technical language detection patternsThis is expected behavior; suggestions only appear when TECHNICAL_PATTERNS match entry descriptions

Success Criteria

  • Every release has structured notes published within 24 hours of deployment
  • All entries lead with user benefit, not technical implementation details
  • Breaking changes include explicit migration instructions with deadlines
  • Deprecations include removal timeline and recommended alternatives
  • Technical language flags reviewed and addressed before publication
  • Release notes tone matches the target audience (B2B/consumer/developer/internal)
  • Entries traceable to tickets via ticket_id for audit and context

Scope & Limitations

In Scope:

  • Structured release note generation from JSON input with 5 entry categories
  • Automatic technical language detection with rewriting suggestions
  • Markdown and JSON output formatting with category grouping
  • Audience tone guidance (B2B, consumer, developer, internal)
  • Classification rules for New Features, Improvements, Bug Fixes, Breaking Changes, and Deprecations

Out of Scope:

  • Automatic extraction of changes from git history (see sprint-retrospective/ for git analysis)
  • Jira/Linear ticket integration for pulling completed stories (manual JSON input required)
  • Changelog maintenance across multiple releases (this tool generates per-release notes)
  • Distribution to email, in-app, or documentation channels (output is markdown/JSON for further processing)

Important Caveats:

  • The rewriting suggestions are pattern-based heuristics. They catch common technical language but cannot assess whether a description truly communicates user benefit.
  • Semantic versioning alignment is the user's responsibility. The tool does not validate that version numbers follow semver conventions relative to the change types present.
  • Breaking changes require special care. Always include: what changed, what the user must do, and by when. Vague breaking change notes erode user trust.

Integration Points

IntegrationDirectionDescription
sprint-retrospective/Receives fromSprint commit data and type distribution inform what changes to include
senior-pm/ComplementsStakeholder communication plans guide release note audience and tone
execution/create-prd/Receives fromPRD feature descriptions (Section 7) become release note entry drafts
scrum-master/Receives fromSprint review outputs identify what shipped and needs documentation
summarize-meeting/Receives fromRelease planning meeting summaries capture context for release notes
job-stories/ / wwas/Receives fromUser story descriptions inform user-benefit framing of entries

Tool Reference

release_notes_generator.py

Generates formatted release notes from structured JSON input. Groups entries by category, formats into markdown or JSON, and flags entries that may need user-benefit rewriting.

FlagTypeDefaultDescription
--inputstring(optional)Path to JSON file containing release entries
--demoflagoffRun with built-in demo data (8 entries across all types)
--product-namestring(required)Product name for the release notes header
--versionstring(required)Version string (e.g., 2.5.0)
--formatchoicetextOutput format: text (markdown) or json
--datestringtodayRelease date in YYYY-MM-DD format

Input JSON schema:

{
  "entries": [
    {
      "title": "Feature Name",
      "description": "User-benefit description (1-3 sentences)",
      "type": "feature|improvement|bugfix|breaking|deprecation",
      "ticket_id": "PROJ-123 (optional)"
    }
  ]
}

References

  • See references/release-notes-guide.md for best practices, audience guidance, and examples.
  • See assets/release_notes_template.md for a ready-to-use document template.

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

35.83%
按下载量换算170

Claude

31.74%
按下载量换算151

Cursor

20.22%
按下载量换算96

Gemini CLI

9.5%
按下载量换算45

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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。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