MCPInstall

semantic-release

Automate versioning with Node.js semantic-release v25+. TRIGGERS - npm run release, version bump, changelog, conventional commits, release automation.

Installs
213
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/terrylica/cc-skills --skill semantic-release
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads semantic-release automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

semantic-release

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

Overview

Automate semantic versioning and release management using semantic-release v25+ (Node.js) following 2025 best practices. Works with all languages (JavaScript, TypeScript, Python, Rust, Go, C++, etc.) via the @semantic-release/exec plugin. Create shareable configurations for multi-repository setups, initialize individual projects with automated releases, and configure GitHub Actions workflows with OIDC trusted publishing.

Important: This skill uses semantic-release (Node.js) exclusively, NOT python-semantic-release, even for Python projects. Rationale: 23.5x larger community, 100x+ adoption, better future-proofing.

When to Use This Skill

Invoke when:

  • Setting up local releases for a new project (any language)
  • Creating shareable semantic-release configuration for organization-wide use
  • Migrating existing projects to 2025 semantic-release patterns
  • Troubleshooting semantic-release setup or version bumps
  • Setting up Python projects (use Node.js semantic-release, NOT python-semantic-release)
  • Configuring GitHub Actions (optional backup, not recommended as primary due to speed)
  • Rust workspaces using release-plz (see Rust reference)

Why Node.js semantic-release

22,900 GitHub stars - Large, active community 1.9M weekly downloads - Proven adoption 126,000 projects using it - Battle-tested at scale 35+ official plugins - Rich ecosystem Multi-language support - Works with any language via @semantic-release/exec

Do NOT use python-semantic-release. It has a 23.5x smaller community (975 vs 22,900 stars), ~100x less adoption, and is not affiliated with the semantic-release organization.


Release Workflow Philosophy: Local-First

Default approach: Run releases locally, not via GitHub Actions.

Why Local Releases

Primary argument: GitHub Actions is slow

  • ⏱️ GitHub Actions: 2-5 minute wait for release to complete
  • ⚡ Local release: Instant feedback and file updates
  • 🔄 Immediate workflow continuity - no waiting for CI/CD

Additional benefits:

  • ✅ Instant local file sync - package.json, CHANGELOG.md, tags updated immediately
  • ✅ No pull required - Continue working without git pull after release
  • ✅ Dry-run testing - npm run release:dry to preview changes before release
  • ✅ Offline capable - Can release without CI/CD dependency
  • ✅ Faster iteration - Debug release issues immediately, not through CI logs

GitHub Actions: Optional Backup Only

GitHub Actions workflows are provided as optional automation, not the primary method:

  • Use for team consistency if required
  • Backup if local environment unavailable
  • Not recommended as primary workflow due to speed

Authentication Setup

gh auth login
# Browser authentication once
# Credentials stored in keyring
# All future releases: zero manual intervention

This is the minimum manual intervention possible for local semantic-release with GitHub plugin functionality.

Multi-Account Authentication via mise [env]

For multi-account GitHub setups, use mise [env] to set per-directory GH_TOKEN:

# ~/your-project/.mise.toml
[env]
GH_TOKEN = "{{ read_file(path=env.HOME ~ '/.claude/.secrets/gh-token-accountname') | trim }}"
GITHUB_TOKEN = "{{ read_file(path=env.HOME ~ '/.claude/.secrets/gh-token-accountname') | trim }}"

This overrides gh CLI's global authentication, ensuring semantic-release uses the correct account for each directory.

See the mise-configuration skill for complete setup.

mise Task Detection

When .mise.toml has release tasks, prefer mise run over npm run:

PriorityConditionCommand
1.mise.toml has [tasks.release:*]mise run release:version
2package.json has scripts.releasenpm run release
3Global semantic-releasesemantic-release --no-ci

See Python Guide for complete mise workflow example.

GitHub Actions Policy

CRITICAL: No testing or linting in GitHub Actions. See CLAUDE.md for full policy.

ForbiddenAllowed
pytest, npm test, cargo testsemantic-release
ruff, eslint, clippy, prettierCodeQL, npm audit
mypyDeployment, Dependabot

Separation of Concerns (4-Level Architecture)

semantic-release configuration follows a hierarchical, composable pattern:

Level 1: Skill - ${CLAUDE_PLUGIN_ROOT}/skills/semantic-release/ (Generic templates, system-wide tool) Level 2: User Config - ~/semantic-release-config/ (@username/semantic-release-config) Level 3: Organization Config - npm registry (@company/semantic-release-config) Level 4: Project Config - .releaserc.yml in project root

Configuration Precedence

Level 4 (Project) → overrides → Level 3 (Org) → overrides → Level 2 (User) → overrides → Defaults

Conventional Commits Format

semantic-release analyzes commit messages to determine version bumps:

<type>(<scope>): <subject>

Version Bump Rules (Default)

  • feat: → MINOR version bump (0.1.0 → 0.2.0)
  • fix: → PATCH version bump (0.1.0 → 0.1.1)
  • BREAKING CHANGE: or feat!: → MAJOR version bump (0.1.0 → 1.0.0)
  • docs:, chore:, style:, refactor:, perf:, test: → No version bump (by default)

Release Notes Visibility (Important)

Warning: The @semantic-release/release-notes-generator (Angular preset) only includes these types in release notes:

  • feat: → Features section
  • fix: → Bug Fixes section
  • perf: → Performance Improvements section

Other types (docs:, chore:, refactor:, etc.) trigger releases when configured but do NOT appear in release notes.

Recommendation: For documentation changes that should be visible in release notes, use:

fix(docs): description of documentation improvement

This ensures the commit appears in the "Bug Fixes" section while still being semantically accurate (fixing documentation gaps is a fix).

Marketplace Plugin Configuration (Always Bump)

For Claude Code marketplace plugins, every change requires a version bump for users to receive updates.

Option A: Shareable Config (if published)

# .releaserc.yml
extends: "@terryli/semantic-release-config/marketplace"

Option B: Inline Configuration

# .releaserc.yml
plugins:
  - - "@semantic-release/commit-analyzer"
    - releaseRules:
        # Marketplace plugins require version bump for ANY change
        - { type: "docs", release: "patch" }
        - { type: "chore", release: "patch" }
        - { type: "style", release: "patch" }
        - { type: "refactor", release: "patch" }
        - { type: "test", release: "patch" }
        - { type: "build", release: "patch" }
        - { type: "ci", release: "patch" }

Result after configuration:

Commit TypeRelease Type
feat:minor (default)
fix:, perf:, revert:patch (default)
docs:, chore:, style:, refactor:, test:, build:, ci:patch (configured)

Why marketplace plugins need this: Plugin updates are distributed via version tags. Without a version bump, users running /plugin update see no changes even if content was modified.

MANDATORY: Every Release Must Increment Version

Pre-release validation: Before running semantic-release, verify releasable commits exist since last tag. A release without version increment is invalid.

Autonomous check sequence:

  1. List commits since last tag: compare HEAD against latest version tag
  2. Identify commit types: scan for feat:, fix:, or BREAKING CHANGE: prefixes
  3. If NO releasable commits found → STOP - do not proceed with release
  4. Inform user: "No version-bumping commits since last release. Use feat: or fix: prefix for releasable changes."

Commit type selection guidance:

  • Use fix: for any change that improves existing behavior (bug fixes, enhancements, documentation corrections that affect usage)
  • Use feat: for new capabilities or significant additions
  • Reserve chore:, docs:, refactor: for changes that truly don't warrant a release

Why this matters: A release without version increment creates confusion - users cannot distinguish between releases, package managers may cache old versions, and changelog entries become meaningless.

MAJOR Version Breaking Change Confirmation

Trigger: BREAKING CHANGE: footer or feat!: / fix!: prefix in commits.

When MAJOR is detected, this skill runs a 3-phase confirmation workflow:

  1. Detection: Scan commits for breaking change markers
  2. Analysis: Spawn 3 parallel subagents (User Impact, API Compat, Migration)
  3. Confirmation: AskUserQuestion with proceed/downgrade/abort options

See MAJOR Confirmation Workflow for complete details including subagent prompts, decision tree, and example output.

Examples

Feature (MINOR):

feat: add BigQuery data source support

Bug Fix (PATCH):

fix: correct timestamp parsing for UTC offsets

Breaking Change (MAJOR):

feat!: change API to require authentication

BREAKING CHANGE: All API calls now require API key in Authorization header.

Documentation Linking

Auto-include doc changes in release notes. Add to .releaserc.yml:

- - "@semantic-release/exec"
  - generateNotesCmd: "node plugins/itp/skills/semantic-release/scripts/generate-doc-notes.mjs ${lastRelease.gitTag}"

Detects: ADRs, Design Specs, Skills, Plugin READMEs. See Doc Release Linking.

Note: The @semantic-release/exec plugin uses Lodash templates (${var}). This conflicts with bash default syntax (${VAR:-default}) and subshell syntax ($(cmd)). Preferred fix: remove successCmd entirely if your task runner already handles post-release steps. See Troubleshooting: Lodash Template Conflicts.


Quick Start

Prerequisites

CheckCommandFix
gh CLI authenticatedgh auth statusgh auth login
GH_TOKEN for directorygh api user --jq '.login'See Authentication
Git remote is HTTPSgit remote get-url origingit-ssh-to-https
semantic-release globalcommand -v semantic-releaseSee Troubleshooting

Initialize Project

./scripts/init-project.mjs --project   # Initialize current project
./scripts/init-project.mjs --user      # Create user-level shareable config
./scripts/init-project.mjs --help      # See all options

Run Release

PriorityConditionCommands
1.mise.toml has release tasksmise run release:version / mise run release:full
2package.json has scriptsnpm run release:dry (preview) / npm run release
3Global CLIsemantic-release --no-ci

See Local Release Workflow for the complete 4-phase process.

Python Projects

semantic-release handles versioning. For PyPI publishing, see pypi-doppler skill.

Version pattern (importlib.metadata - never hardcode):

from importlib.metadata import PackageNotFoundError, version
try:
    __version__ = version("your-package-name")
except PackageNotFoundError:
    __version__ = "0.0.0+dev"

See Python Projects Guide for complete setup including Rust+Python hybrids.

GitHub Actions (Optional)

Not recommended as primary (2-5 minute delay). Repository Settings → Actions → Workflow permissions → Enable "Read and write permissions".


Reference Documentation

CategoryReferenceDescription
SetupAuthenticationHTTPS-first setup, multi-account patterns
WorkflowLocal Release Workflow4-phase process (PREFLIGHT → RELEASE → POSTFLIGHT)
LanguagesPython ProjectsPython + Rust+Python hybrid patterns
Rust Projectsrelease-plz, cargo-rdme README SSoT
ConfigVersion AlignmentGit tags as SSoT, manifest patterns
Monorepo SupportPolyglot monorepo with Pants + mise, pnpm/npm workspaces
AdvancedMAJOR ConfirmationBreaking change analysis workflow
Doc Release LinkingAuto-link ADRs/specs in release notes
HelpTroubleshootingAll common issues consolidated
Evolution LogSkill change history

Cross-skill references:

  • mise-tasks skill: polyglot-affected - Complete Pants + mise integration guide
  • mise-tasks skill: bootstrap-monorepo - Autonomous polyglot monorepo setup
  • pypi-doppler skill - Local PyPI publishing with Doppler

Post-Change Checklist

After modifying THIS skill (semantic-release):

  1. SKILL.md and references remain aligned
  2. New references documented in Reference Documentation table
  3. All referenced files in references/ exist
  4. Append changes to evolution-log.md
  5. Validate with bun scripts/validate-plugins.mjs
  6. Run npm run release:dry to verify no regressions

Troubleshooting

IssueCauseSolution
No release createdNo releasable commits since tagUse feat: or fix: prefix for version-bumping commits
Wrong version bumpCommit type mismatchCheck conventional commit format and releaseRules
GitHub release not createdMissing GH_TOKEN or permissionsCheck token is set and has repo scope
CHANGELOG not updatedMissing changelog pluginAdd @semantic-release/changelog to plugins array
"Authentication failed"HTTPS vs SSH remote mismatchConvert to HTTPS: git-ssh-to-https
semantic-release not foundNot installed globallynpm install -g semantic-release
Gatekeeper blocks on macOSUnsigned Node filesSee Troubleshooting
dry-run shows no changesAlready released at HEADMake new commits before running release
Multi-account token conflictWrong GH_TOKEN for directoryConfigure mise [env] per-directory token

Post-Execution Reflection

After this skill completes, check before closing:

  1. Did the command succeed? — If not, fix the instruction or error table that caused the failure.
  2. Did parameters or output change? — If the underlying tool's interface drifted, update Usage examples and Parameters table to match.
  3. Was a workaround needed? — If you had to improvise (different flags, extra steps), update this SKILL.md so the next invocation doesn't need the same workaround.

Only update if the issue is real and reproducible — not speculative.

FAQ

What does the semantic-release skill do?

Automate versioning with Node.js semantic-release v25+. TRIGGERS - npm run release, version bump, changelog, conventional commits, release automation.

How do I install semantic-release?

Run npx skills add https://github.com/terrylica/cc-skills --skill semantic-release in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/semantic-release/ (personal) or .claude/skills/semantic-release/ (project).

Which AI agents can use semantic-release?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is semantic-release safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains semantic-release?

It is published by terrylica in terrylica/cc-skills. It has 213 recorded installs.

More from terrylica/cc-skills

View all →

mise-tasks

terrylica/cc-skills

Orchestrate multi-step project workflows using mise task definitions with dependency management and argument handling. Use whenever the user.

pandoc-pdf-generation

terrylica/cc-skills

PDF generation from markdown via Pandoc/XeLaTeX. TRIGGERS - markdown for PDF, print document, pandoc

Docs & Writing547 installs

mql5-indicator-patterns

terrylica/cc-skills

MQL5 indicator development patterns. TRIGGERS - MQL5 indicator, OnCalculate, indicator buffers, MetaTrader 5.

501 installs

Python logging with loguru, structlog, and orjson. TRIGGERS - loguru, structlog, structured logging

DevOps & Cloud402 installs

gdrive-access

terrylica/cc-skills

Open the skill to read its SKILL.md instructions.

356 installs

chezmoi-workflows

terrylica/cc-skills

Open the skill to read its SKILL.md instructions.

297 installs

Popular productivity & automation skills

View category →

agent-browser

vercel-labs/agent-browser

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.

OfficialProductivity & Automation1.3M installs

lark-doc

open.feishu.cn

飞书云文档(Docx / Wiki)内容操作:读取、创建、编辑文档,插入或下载图片附件,以及操作思维笔记。用户提供文档 URL/token(包括 doubao.com 的 /docx/、/wiki/)时使用;按 URL 路径/token 而非域名路由;/page/ 是妙搭应用,非飞书云文档。文档内嵌资源按读取参考中的统一规则分流。独立评论操作走 lark-drive;随正文读取评论使用 docs +fetch。表格或 Base 内部数据操作不在本 skill。

lark-base

open.feishu.cn

飞书多维表格(Base)操作:建表、字段、记录、视图、统计、公式/lookup、表单、仪表盘、应用模式(BaseApp/AppMode 页面与组件)、Workspace 目录、workflow、角色权限、模板中心(多维表格模板分类/列表/搜索);遇到 Base/多维表格/bitable、BaseApp/AppMode、/base/ 或 /app/ 链接时使用。BaseApp 不走 lark-apps;文件导入/导出转 lark-drive,认证/授权转 lark-shared。

lark-drive

open.feishu.cn

飞书云空间(云盘/云存储):管理 Drive 文件和文件夹,包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、查询权限设置、评论/权限/订阅、标题、版本、飞书文档密级标签(secure labels)和本地文件导入。用户需要整理云盘目录、处理云空间资源 URL/token、判断链接类型/真实 token/标题,或导入 Word/Markdown/Excel/CSV/PPTX/.base 为 docx/sheet/bitable/slides 时使用;doubao.com 云空间 URL/token 也按资源路径和 token 路由,不回退 WebFetch。不负责:文档内容编辑(走 lark-doc)、表格/Base 表内数据操作(走 lark-sheets/lark-base)、知识空间节点/成员管理(走 lark-wiki)、原生 Markdown 文件读写/patch/diff(走 lark-markdown)。

lark-shared

open.feishu.cn

Use for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions (--domain, including all/docs/drive), missing scopes, revoking authorization, or handling notice JSON.

lark-im

open.feishu.cn

飞书即时通讯:收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片(Interactive Card)、监听卡片按钮回调(card.action.trigger)。当用户需要发消息、查看或搜索聊天记录、下载聊天中的文件、查看群成员、搜索群、创建群聊或话题群、管理标记数据、管理 Feed 置顶(添加/移除/查询置顶会话)、管理标签数据、处理卡片回调时使用。