Agent skill
push
Create/update GitHub issue, branch, commit, PR, then continue implementing
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/push-alexmond-jhelm
SKILL.md
Push Workflow: Issue → Branch → PR → Continue
Argument $ARGUMENTS is either:
- An existing issue number (e.g.
42) — update it and wire up the branch/PR - A short description (e.g.
add retry logic for HTTP calls) — create a new issue first
Step 1: Resolve or create the GitHub issue
If $ARGUMENTS is a number — view the existing issue:
gh issue view $ARGUMENTS
If $ARGUMENTS is a description — create a new issue:
gh issue create \
--title "$ARGUMENTS" \
--body "## Context
<fill in context from current conversation / recent commits>
## Acceptance criteria
- [ ] <criterion 1>
- [ ] <criterion 2>"
Capture the new issue number from the output (e.g. #59).
Step 1.5: Shortcut — skill-only changes go directly to main
If the only changed files are under .claude/skills/ (skill updates, no source code changes):
git diff --stat
If all changed files are .claude/skills/** files:
- Skip Steps 2–6 (no branch, no PR)
- Commit directly on
main:bashgit add .claude/skills/<changed-file> git commit -m "$(cat <<'EOF' <imperative summary under 72 chars> Co-Authored-By: Claude Sonnet 4.6 <[email protected]> EOF )" git push - Report the commit SHA and stop. No CI wait needed for skill-only commits.
Otherwise continue with the full workflow below.
Step 2: Ensure we're on main and up to date
git checkout main && git pull
Step 3: Create a feature branch named after the issue
git checkout -b feature/<issue-number>-<short-slug>
Where <short-slug> is 2–4 words from the issue title, kebab-cased (e.g. feature/59-retry-http-calls).
If a matching branch already exists, check it out instead:
git checkout feature/<issue-number>-<short-slug>
Step 4: Stage and commit any local changes (if present)
Check for uncommitted work first:
git status
git diff --stat
If there are changes to commit:
git add <changed-files>
git commit -m "$(cat <<'EOF'
<imperative summary under 72 chars>
Closes #<issue-number>
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
EOF
)"
If there are no changes yet, skip this step — implementation comes after the PR is open.
Step 5: Push the branch
git push -u origin HEAD
Step 6: Create the pull request (draft if no code yet)
gh pr create \
--title "<concise title matching issue>" \
--body "$(cat <<'EOF'
## Summary
- <bullet 1>
- <bullet 2>
## Test plan
- [ ] Run `./mvnw test -pl <module>`
- [ ] Verify <key behavior>
Closes #<issue-number>
🤖 Generated with [Claude Code](https://claude.com/claude-code)
EOF
)" \
--draft
Remove --draft if there are already commits with real code on the branch.
Report the PR URL to the user.
Step 7: Continue implementing
Now implement the work described in the issue:
- Read the issue body for full requirements
- Follow project coding standards (Java 21, Lombok, tabs, spring-javaformat)
- Run
./mvnw test -pl <module>after each meaningful change - When done, mark the PR ready for review:
bash
gh pr ready
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?