Agent skill
gitflow-workflow
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/gitflow-workflow
SKILL.md
Purpose
All detailed command manuals live under ./references/ and should be consulted when you need exact flags/options.
Required invariants (used by the 6 instruction-type skills)
Preflight
Before any start/finish operation:
- Confirm working tree is clean (use
git statusto verify). - Identify current branch and verify the branch type matches the operation (
feature/*,hotfix/*,release/*). - Determine the active workflow preset (Classic GitFlow vs GitHub Flow vs GitLab Flow) to identify correct base/target branches.
Branch naming
- Use kebab-case and a strict prefix:
<type>/<kebab-case>(e.g.feature/user-authentication).
Finish behavior
- Prefer non-fast-forward merges when finishing (e.g.
--no-ff) to preserve a visible integration history, unless the repo workflow config says otherwise. - Run tests if available before finishing; fix failures before merging.
- After finishing hotfix/release, ensure the integration branch receives the changes (often merging back into
develop). - Delete topic branches locally and remotely after finishing when policy allows.
Changelog generation
When updating CHANGELOG.md during finish-release or finish-hotfix operations, follow these principles:
Core principle: Include all commits that represent user-facing changes since the previous version tag. Exclude internal-only changes (build, CI, tooling) and administrative commits (merges, releases).
Note: The finish-feature operation updates the [Unreleased] section without filtering, as features are already user-facing by nature.
Commit collection process for releases and hotfixes:
- Identify the previous version tag (e.g.,
v1.5.0) - Collect commits:
git log --oneline <previous-tag>..HEAD - Filter commits using conventional commit format, applying the user-facing principle:
- Include: Commits that affect end users, API consumers, or documented behavior (e.g.,
feat,fix,refactor,docs,perf,deprecate,remove,security) - Exclude: Internal tooling (
chore,build,ci), test-only changes (test), merge commits, release commits
- Include: Commits that affect end users, API consumers, or documented behavior (e.g.,
- Map commit types to appropriate Keep a Changelog sections (
Added,Changed,Deprecated,Removed,Fixed,Security) - Format entries using present/imperative tense and user-facing language
- Merge repetitive items and include sufficient context to understand impact
See ${CLAUDE_PLUGIN_ROOT}/examples/changelog.md for the standard format.
Minimal templates (what the 6 skills assume)
The operation-specific templates have been moved into the corresponding start-* / finish-* skills to keep those tasks self-contained.
Reference index
./references/context-gathering.md./references/naming-rules.md./references/workflow-presets.md./references/versioning-and-tags.md./references/commands-core.md./references/commands-topic.md
External References
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?