Finishing a development branch
Skill stanlrt/cc-superpowers-parallelised/skills/finishing-a-development-branch
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanupFrom its SKILL.md
npx -y skills add stanlrt/cc-superpowers-parallelised --skill finishing-a-development-branchAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
11.7 KB, as published. Nobody here has run it
Finishing a Development Branch
Overview
Guide completion of development work by presenting clear options and handling chosen workflow.
Core principle: Verify tests → Detect environment → Present options → Execute choice → Clean up.
Announce at start: "I'm using the finishing-a-development-branch skill to complete this work."
The Process
Step 1: Verify Tests
Before presenting options, verify tests pass:
# Run project's test suite
npm test / cargo test / pytest / go test ./...
If tests fail:
Tests failing (<N> failures). Must fix before completing:
[Show failures]
Cannot proceed with merge/PR until tests pass.
Stop. Don't proceed to Step 2.
If tests pass: Continue to Step 1b.
Step 1b: Surface Deferred Refactor Items
If subagent-driven-development ran, its whole-branch refactor reviewer may have left design-opinion findings for triage:
cat "$(git rev-parse --show-toplevel)/.superpowers/sdd/refactor-report.md" 2>/dev/null
If the file exists and is non-empty, present each Refactor-Advisory item to your human partner before presenting the completion options, and get a decision per item — fix now (do it, re-run tests, then continue), ticket (record it — e.g. a TODO or issue — and note where), or accept (leave as-is, on the record). Do not merge or open a PR while unaddressed advisory items sit unseen; a report nobody reads is a silent discard. If the file is absent, this skill was reached by another path — continue.
Step 2: Detect Environment
Determine workspace state before presenting options:
GIT_DIR=$(cd "$(git rev-parse --git-dir)" 2>/dev/null && pwd -P)
GIT_COMMON=$(cd "$(git rev-parse --git-common-dir)" 2>/dev/null && pwd -P)
This determines which menu to show and how cleanup works:
| State | Menu | Cleanup |
|---|---|---|
GIT_DIR == GIT_COMMON (normal repo) | Standard 4 options | No worktree to clean up |
GIT_DIR != GIT_COMMON, named branch | Standard 4 options | Provenance-based (see Step 6) |
GIT_DIR != GIT_COMMON, detached HEAD | Reduced 3 options (no merge) | No cleanup (externally managed) |
Step 3: Determine Base Branch
# Try common base branches
git merge-base HEAD main 2>/dev/null || git merge-base HEAD master 2>/dev/null
Or ask: "This branch split from main - is that correct?"
Step 4: Present Options
Normal repo and named-branch worktree — present exactly these 4 options:
Implementation complete. What would you like to do?
1. Merge back to <base-branch> locally
2. Push and create a Pull Request
3. Keep the branch as-is (I'll handle it later)
4. Discard this work
Which option?
Detached HEAD — present exactly these 3 options:
Implementation complete. You're on a detached HEAD (externally managed workspace).
1. Push as new branch and create a Pull Request
2. Keep as-is (I'll handle it later)
3. Discard this work
Which option?
Don't add explanation - keep options concise.
Step 5: Execute Choice
Option 1: Merge Locally
# Get main repo root for CWD safety
MAIN_ROOT=$(git -C "$(git rev-parse --git-common-dir)/.." rev-parse --show-toplevel)
cd "$MAIN_ROOT"
# Merge first — verify success before removing anything
git checkout <base-branch>
git pull
git merge <feature-branch>
# Verify tests on merged result
<test command>
# Only after merge succeeds: cleanup worktree (Step 6), then delete branch
Then: Cleanup worktree (Step 6), then delete branch:
git branch -d <feature-branch>
Option 2: Push and Create PR
Rebase onto the latest base before pushing. A branch cut earlier falls behind while you work — and any PR that merges in the meantime moves the base. CI diffs your branch against the current base, so pushing a stale branch makes the PR show unrelated files as changed (reverts, phantom scope). Fetch and rebase first:
git fetch origin
git rebase origin/<base-branch> # usually main
# resolve any conflicts, then push
git push -u origin <feature-branch>
If the PR then sits open while <base-branch> moves again, repeat
git fetch origin && git rebase origin/<base-branch> and re-push with
--force-with-lease before it can merge.
Build the PR body from the repo's template, then create with --body-file.
-
If
.github/pull_request_template.md(or.github/PULL_REQUEST_TEMPLATE.md, ordocs/pull_request_template.md) exists, base the body on it. Fill every section and tick every required checklist item (- [ ]→- [x]). Required-checklist CI jobs fail an un-ticked box, so leaving the template default unchecked will block the PR. Leave items tagged(optional)unchecked only when they genuinely don't apply. Delete any template callout that instructs the author to delete it (e.g. a title-format warning). If no template exists, fall back to a## Summary+## Test Planbody. -
Write the finished body to a file, then pass it with
--body-file— never inline via--body "..."or a heredoc. Inline bodies break on apostrophes, backticks, and$()in shell, and silently drop the template's checklist.
# Write the filled-in body to a scratch file (use the Write tool, not echo)
gh pr create --title "<title>" --body-file <path-to-body-file>
The title must satisfy the repo's PR-title rule (Conventional Commits where enforced). After creating, delete the scratch body file.
Then monitor CI to completion. Do not guess a flat wait — CI duration varies, so a fixed 3-minute sleep either races the run or wastes time. Poll the PR's checks on a 30-second interval, and when they settle, print any failures in the same pass — don't re-poll by hand to hunt them down:
PR=<pr-number>
while :; do
# One gh call per iteration. gh exits non-zero both when checks are pending and
# when none exist yet, so judge by the output, not the exit code. A real error
# (auth/network/rate-limit) yields no counts and prints to stderr — surface it
# and keep looping, rather than silently spinning on "0 checks".
err=$(mktemp) # private temp; avoids a predictable /tmp path and concurrent clobber
counts=$(gh pr checks "$PR" --json bucket \
-q '"\(length) \([.[] | select(.bucket == "pending")] | length)"' 2>"$err")
if [ -z "$counts" ]; then
grep -qiE 'no checks|no commit statuses' "$err" \
&& echo "No checks registered yet — re-checking in 30s" \
|| { echo "gh pr checks failed — re-checking in 30s:" >&2; cat "$err" >&2; }
rm -f "$err"; sleep 30; continue
fi
rm -f "$err"
total=${counts%% *}; pending=${counts##* } # portable split (works in bash and zsh)
[ "$total" -gt 0 ] && [ "$pending" -eq 0 ] && break
echo "CI not settled yet ($total checks, $pending pending) — re-checking in 30s"
sleep 30
done
# Settled. If anything failed, dump its logs right here — no second poll.
if gh pr checks "$PR"; then
echo "CI GREEN"
else
gh pr checks "$PR" --json bucket,link -q '.[] | select(.bucket=="fail") | .link' \
| grep -oE '/runs/[0-9]+' | grep -oE '[0-9]+' | sort -u \
| while read -r run; do echo "=== failed run $run ==="; gh run view "$run" --log-failed; done
fi
(If your shell blocks a foreground sleep, run the whole block as one
background command and read its output. Never pipe the status command through
tail/head — $? then reports the pipe's last stage, masking gh's
non-zero exit, and you misread a red run as green.) If the block prints
failures, fix the cause, push, and re-enter it — do not proceed until it
prints CI GREEN.
Once CI is green, assess the Copilot + Code Quality comments. Fix+resolve them, or answer why they aren't relevant+resolve them. To resolve Copilot comments: gh cli. To resolve Code Quality comments: GH GraphQL API via CLI.
Do NOT clean up worktree — user needs it alive to iterate on PR feedback.
Option 3: Keep As-Is
Report: "Keeping branch <name>. Worktree preserved at <path>."
Don't cleanup worktree.
Option 4: Discard
Confirm first:
This will permanently delete:
- Branch <name>
- All commits: <commit-list>
- Worktree at <path>
Type 'discard' to confirm.
Wait for exact confirmation.
If confirmed:
MAIN_ROOT=$(git -C "$(git rev-parse --git-common-dir)/.." rev-parse --show-toplevel)
cd "$MAIN_ROOT"
Then: Cleanup worktree (Step 6), then force-delete branch:
git branch -D <feature-branch>
Step 6: Cleanup Workspace
Only runs for Options 1 and 4. Options 2 and 3 always preserve the worktree.
GIT_DIR=$(cd "$(git rev-parse --git-dir)" 2>/dev/null && pwd -P)
GIT_COMMON=$(cd "$(git rev-parse --git-common-dir)" 2>/dev/null && pwd -P)
WORKTREE_PATH=$(git rev-parse --show-toplevel)
If GIT_DIR == GIT_COMMON: Normal repo, no worktree to clean up. Done.
If worktree path is under .worktrees/ or worktrees/: Superpowers created this worktree — we own cleanup.
MAIN_ROOT=$(git -C "$(git rev-parse --git-common-dir)/.." rev-parse --show-toplevel)
cd "$MAIN_ROOT"
git worktree remove "$WORKTREE_PATH"
git worktree prune # Self-healing: clean up any stale registrations
Otherwise: The host environment (harness) owns this workspace. Do NOT remove it. If your platform provides a workspace-exit tool, use it. Otherwise, leave the workspace in place.
Quick Reference
| Option | Merge | Push | Keep Worktree | Cleanup Branch |
|---|---|---|---|---|
| 1. Merge locally | yes | - | - | yes |
| 2. Create PR | - | yes | yes | - |
| 3. Keep as-is | - | - | yes | - |
| 4. Discard | - | - | - | yes (force) |
Common Mistakes
Skipping test verification
- Problem: Merge broken code, create failing PR
- Fix: Always verify tests before offering options
Open-ended questions
- Problem: "What should I do next?" is ambiguous
- Fix: Present exactly 4 structured options (or 3 for detached HEAD)
Cleaning up worktree for Option 2
- Problem: Remove worktree user needs for PR iteration
- Fix: Only cleanup for Options 1 and 4
Deleting branch before removing worktree
- Problem:
git branch -dfails because worktree still references the branch - Fix: Merge first, remove worktree, then delete branch
Running git worktree remove from inside the worktree
- Problem: Command fails silently when CWD is inside the worktree being removed
- Fix: Always
cdto main repo root beforegit worktree remove
Cleaning up harness-owned worktrees
- Problem: Removing a worktree the harness created causes phantom state
- Fix: Only clean up worktrees under
.worktrees/orworktrees/
No confirmation for discard
- Problem: Accidentally delete work
- Fix: Require typed "discard" confirmation
Red Flags
Never:
- Create a PR with an inline
--body/heredoc, or ignore the repo's PR template (breaks on shell metacharacters; drops the required checklist → CI fails) - Proceed with failing tests
- Merge without verifying tests on result
- Delete work without confirmation
- Force-push without explicit request
- Remove a worktree before confirming merge success
- Clean up worktrees you didn't create (provenance check)
- Run
git worktree removefrom inside the worktree
Always:
- Verify tests before offering options
- Detect environment before presenting menu
- Present exactly 4 options (or 3 for detached HEAD)
- Get typed confirmation for Option 4
- Clean up worktree for Options 1 & 4 only
cdto main repo root before worktree removal- Run
git worktree pruneafter removal