Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Rebuild Branch · platformplatform | DecimalAI

codingby platformplatform

Rebuild Branch

platformplatform/rebuild-branch

Source platformplatform/platformplatform ↗License MIT

Rebuild a stale branch by cherry-picking each commit onto a fresh branch off main, using a ralph-loop to validate each commit (build, test, format, lint, optional e2e) before moving on.

SkillScore39Maintained

Automatic

Coding pack

Last verified 8/20/2026

What it changes

Passed 4 of 10 test cases with the skill, and 1 of 10 without it.

Without skill13%

With skill39%

Checks it passed that the model alone missed

Orders validation sequence with build first
Orders validation sequence with build step first
Removes comment lines starting with hash symbols

and 10 more in the full report

8 fixed1 passed either way12 still failing2 broke

Pass-rate lift: +26 pp
Tokens: +69%
Agent turns: 0%
Response time: -57%

Model gemini-3.6-flashMeasured 8/20/2026v19 of 23 cases passed with skill · 10 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/platformplatform-rebuild-branch/SKILL.md https://app.decimal.ai/s/platformplatform-rebuild-branch/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add platformplatform-rebuild-branch, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/platformplatform-rebuild-branch/ and pins it in .decimal/skills.lock

decimalai skills add platformplatform-rebuild-branch

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/platformplatform-rebuild-branch/SKILL.md https://app.decimal.ai/s/platformplatform-rebuild-branch/SKILL.md

Raw file: /s/platformplatform-rebuild-branch/SKILL.md ↗ · machine-readable summary at /s/platformplatform-rebuild-branch.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 8

Prompt

What step ordering is mandated in the generated loop instructions when a cherry-pick applies cleanly without conflicts?

What was checked, without and with the skill
Checked	Without skill	With skill
Orders validation sequence with build step first	Missed	Passed
Executes backend tests, format, lint, Aspire restart, and e2e tests in parallel for clean cherry-picks	Missed	Passed

case-14 · gemini-3.6-flash · 8/20/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md170 linessingle file

Rebuild Branch

Drive the discovery + planning phase of a branch rebuild. Emit a /ralph-loop:ralph-loop prompt the user runs themselves to execute the cherry-picks one at a time.

STEP 1: Determine source branch

The user may have named the source branch as the skill argument. If not, ask which branch to rebuild.

If the source branch isn't crystal-clear (multiple matches, ambiguous shorthand), confirm the resolved name with the user before proceeding.

STEP 2: Detect resume vs. fresh start

Compute the root repo path (the main worktree, not the current one):

ROOT_REPO=$(dirname "$(git rev-parse --path-format=absolute --git-common-dir)")
CURRENT_BRANCH=$(git branch --show-current)

A run is in progress if $ROOT_REPO/.workspace/$CURRENT_BRANCH/commits.md exists AND the current branch ends with -rebuild.

If resuming:

Read commits.md and count [x] vs [ ] lines. Identify the next pending commit.
Read learnings.md and surface the last 3 entries.
Show the user a status panel: branch name, N of M commits done, next commit hash + title, last learnings.
Use AskUserQuestion to confirm continue and offer a run-mode change. Options: keep current mode, switch to autonomous, switch to interview-upfront, switch to interview-on-problem.
Skip to STEP 7 (emit ralph-loop) with remaining iteration count.

If fresh: continue to STEP 3.

STEP 3: Verify clean working tree

git status --porcelain

If there are uncommitted changes, stop and ask the user to stash or commit first. Do not proceed.

STEP 4: Discover commits

Track record

Benchmark

4 / 10

test cases passed, vs 1 / 10 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

platformplatform / platformplatform

.claude/skills/rebuild-branch/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Rebuild Branch

STEP 1: Determine source branch

STEP 2: Detect resume vs. fresh start

STEP 3: Verify clean working tree

STEP 4: Discover commits

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills