Token导航 LogoToken导航

Agent Skill · 研究检索

release-orchestrator(发布协调器)

release-orchestrator 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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。

release-orchestrator 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 42,785。

总安装 1,042周安装 43GitHub Stars 103下载 341核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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-orchestrator(发布协调器)
来源仓库:https://github.com/borghei/claude-skills
仓库路径:skills/release-orchestrator
安装命令:
npx skills add https://github.com/borghei/claude-skills --skill release-orchestrator
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-orchestrator

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

release-orchestrator 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。它可协调多步骤发布流程中的各项任务,支持自动化触发检查、测试和部署环节。

通过 npx skills add 命令从指定仓库安装,具体用法请结合原始 README 进一步确认。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操作。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。

来源仓库内路径:skills/release-orchestrator

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/borghei/claude-skills --skill release-orchestr…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

release-orchestrator 怎么安装?

在终端执行 npx skills add https://github.com/borghei/claude-skills --skill release-orchestrator。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

release-orchestrator 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

release-orchestrator 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

release-orchestrator 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

release-orchestrator 用的人多吗?

公开数据是累计安装 1,042 次,近一周安装 43 次,下载 341 次,GitHub 103 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

release-orchestrator 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
release-orchestrator

发布协调器 · 本页

1,042103需要联网未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Release Orchestrator

The agent runs pre-flight validation, generates changelogs from conventional commits, auto-bumps semantic versions, and scores deployment readiness with a GO/CONDITIONAL/NO-GO decision.


Quick Start

# Pre-flight: branch sync, secrets, conflicts, commits, deps
python scripts/preflight_checker.py --repo . --base main --verbose

# Changelog from conventional commits
python scripts/changelog_generator.py --repo . --from v1.2.0 --to HEAD

# Auto-detect version bump from commit history
python scripts/version_bumper.py --repo . --dry-run

# Score deployment readiness (7 categories, weighted)
python scripts/release_readiness_scorer.py --input release_data.json --json

Tools Overview

ToolInputOutput
preflight_checker.pyRepo path + base branchPass/fail on 7 checks (sync, conflicts, secrets, commits, deps)
changelog_generator.pyGit repo + ref rangeKeep a Changelog markdown with commit grouping
version_bumper.pyRepo pathNext semver from commit analysis; updates version files
release_readiness_scorer.pyRelease data JSONScore 0-100, GO/CONDITIONAL/NO-GO decision

All tools support --json for machine output. Exit code 0 = pass, 1 = fail (CI-friendly).


Workflow 1: Pre-Flight Validation

python scripts/preflight_checker.py --repo . --base main --json

The agent runs seven automated checks:

  1. Branch sync -- local branch up to date with remote base
  2. Merge conflicts -- dry-run merge to detect conflicts
  3. Uncommitted changes -- fail if working tree is dirty
  4. Secret scanning -- pattern-match for API keys, tokens, passwords (AWS, GCP, GitHub, Stripe, JWT)
  5. Gitignore validation -- .env, credential files covered
  6. Conventional commits -- recent commits follow type(scope): description
  7. Dependency audit -- lock file consistency (package-lock.json, poetry.lock, etc.)

Validation checkpoint: All 7 checks pass. Exit code 0.


Workflow 2: Version Management and Changelog

Step 1 -- Auto-detect version bump.

python scripts/version_bumper.py --repo . --dry-run --json
Commit TypeBumpExample
fix:PATCH (0.0.x)fix(auth): handle expired tokens
feat:MINOR (0.x.0)feat(api): add pagination
feat!: or BREAKING CHANGEMAJOR (x.0.0)feat!: redesign auth flow
docs:, chore:, test:No bumpdocs: update README

Reads from: package.json, pyproject.toml, setup.py, setup.cfg, Cargo.toml, VERSION file. Pre-release support: --pre alpha|beta|rc produces 1.3.0-rc.1.

Step 2 -- Generate changelog.

python scripts/changelog_generator.py --repo . --from latest --to HEAD --output CHANGELOG.md --full

Groups commits by type (Added, Changed, Fixed, Security, Breaking Changes) with hashes and @author attribution.

Step 3 -- Apply version bump.

python scripts/version_bumper.py --repo .  # writes to all discovered version files

Validation checkpoint: --dry-run shows expected version. Changelog covers 100% of commits.


Workflow 3: Deployment Readiness

python scripts/release_readiness_scorer.py --input release_data.json --json

The agent scores across 7 weighted categories:

CategoryWeightMeasures
Tests25%Pass rate, coverage, flaky count
Code Quality20%Lint errors, type errors, complexity, duplication
Documentation15%README, API docs, changelog, migration guide
Security15%No secrets, no critical CVEs, SAST clean
Breaking Changes10%Documented, migration path, deprecation notices
Dependencies10%Lock files consistent, no yanked packages
Rollback Plan5%Procedure documented, DB migration reversible, feature flags

Decision thresholds:

ScoreDecisionAction
80-100GOProceed with deployment
60-79CONDITIONALProceed with mitigations documented
0-59NO-GOAddress blockers first

Any single category below 40 triggers a mandatory blocker regardless of overall score.

Validation checkpoint: Score >= 80 (GO). Zero category blockers.


End-to-End Release Pipeline

Chain all workflows into a single automated pipeline:

#!/bin/bash
set -e

# Phase 1: Pre-flight
python scripts/preflight_checker.py --repo . --base main --json > /tmp/preflight.json

# Phase 2: Tests (project-specific)
python -m pytest --cov=src --cov-report=json:coverage.json -v

# Phase 3: Version bump (dry-run)
python scripts/version_bumper.py --repo . --dry-run --json > /tmp/version.json

# Phase 4: Changelog
python scripts/changelog_generator.py --repo . --from latest --to HEAD

# Phase 5: Readiness assessment
python scripts/release_readiness_scorer.py --input release_data.json --json > /tmp/readiness.json
DECISION=$(python -c "import json; print(json.load(open('/tmp/readiness.json'))['decision'])")
echo "Decision: $DECISION"

Non-interactive by default. Blocks on: pre-flight failure, test failure, or NO-GO readiness.


Release Types

TypeBranch PatternBumpNotes
Hotfixhotfix/v1.2.1 from tagPATCHMinimal fix, branches from release tag
PatchStandard flowPATCHAccumulated bug fixes
MinorStandard flowMINORNew features, backward compatible
MajorStandard flowMAJORBreaking changes, needs migration docs
Pre-releaseStandard flow`--pre alpha\beta\rc`1.3.0-alpha.1 for testing

CI/CD Integration

- name: Pre-flight Check
  run: python scripts/preflight_checker.py --repo . --base main --json > preflight.json

- name: Version Bump
  run: python scripts/version_bumper.py --repo . --dry-run --json > version.json

- name: Changelog
  run: python scripts/changelog_generator.py --repo . --from latest --to HEAD --output CHANGELOG.md

- name: Readiness Score
  run: python scripts/release_readiness_scorer.py --input release_data.json

Git hook: python scripts/preflight_checker.py --repo. --base main in .git/hooks/pre-push.


Anti-Patterns

  1. Skipping pre-flight -- secrets ship to production. Always run pre-flight before any release work.
  2. Manual version bumping -- leads to inconsistencies. Let commit history drive the version.
  3. No rollback plan -- every release needs documented rollback (git revert, feature flags, or DB migration down).
  4. Ignoring single-category blockers -- a 95 overall score with 35 Security = NO-GO.
  5. Changelog after release -- generate before tagging so reviewers can validate.

Troubleshooting

ProblemCauseSolution
Pre-flight "HEAD is detached"CI checked out specific commitCheck out a named branch first
Changelog "No commits found"--from ref does not existVerify tag with git tag -l; use --since date range
Version bumper cannot parse versionNon-semver format (e.g., 1.0)Use MAJOR.MINOR.PATCH in all manifest files
Readiness scorer exits 1 despite high scoreSingle category below 40-point blockerCheck BLOCKERS section; fix failing category
Secret scan false positives on test fixturesPattern matches example tokensLines with "example"/"placeholder" are skipped; move fixtures to non-tracked dir

References

GuidePath
Release Engineering Guidereferences/release_engineering_guide.md
Rollback Strategiesreferences/rollback_strategies.md
CI/CD Best Practicesreferences/ci_cd_best_practices.md

Integration Points

SkillIntegration
senior-devopsPipeline stages consume pre-flight and readiness JSON as gates
senior-qaTest results feed Tests category (25% weight)
senior-secopsSecret scan and CVE counts feed Security category (15%)
code-reviewerCode quality metrics feed Code Quality category (20%)
devops-workflow-engineerWorkflow YAML calls tools as pipeline steps

Last Updated: April 2026 Version: 2.1.0

平台分布

Codex

34.78%
按下载量换算119

Claude

28.48%
按下载量换算97

Cursor

18.59%
按下载量换算63

Gemini CLI

9.2%
按下载量换算31

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills