finishing-a-development-branch

finishing-a-development-branch is a skill for Claude Code, Codex from Crearize/ai-dev-helm. It costs 20 tokens per session (2,490 once invoked), scanned A, a copy of finishing-a-development-branch, MIT.

A process for wrapping up completed coding work after tests pass. It checks the project state and presents ways to integrate or clean up the changes.

In plain words
What is it for?
Use it to run the test suite, identify the repository setup, choose an integration option, and clean up temporary development work when appropriate.
Why use it?
It helps ensure finished work is verified and handled correctly before it is merged, committed, or otherwise closed out.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to run the test suite, identify the repository setup, choose an integration option, and clean up temporary development work when appropriate.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/crearize/ai-dev-helm/finishing-a-development-branch
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Crearize/ai-dev-helm --skill finishing-a-development-branch
Clone the repo
git clone --depth 1 https://github.com/Crearize/ai-dev-helm

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for finishing-a-development-branch

README.md
[![agentmods](https://agentmods.dev/badge/skills/crearize/ai-dev-helm/finishing-a-development-branch/github.svg)](https://agentmods.dev/skills/crearize/ai-dev-helm/finishing-a-development-branch)
Your own site
<a href="https://agentmods.dev/skills/crearize/ai-dev-helm/finishing-a-development-branch"><img src="https://agentmods.dev/badge/skills/crearize/ai-dev-helm/finishing-a-development-branch/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for finishing-a-development-branch

Your own site · 80×15
<a href="https://agentmods.dev/skills/crearize/ai-dev-helm/finishing-a-development-branch"><img src="https://agentmods.dev/badge/skills/crearize/ai-dev-helm/finishing-a-development-branch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 20 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,490 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 100% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00020 $0.02490
Opus 5.5 $0.00008 $0.00996
Sonnet 5.5 $0.00004 $0.00498
Haiku 4.5 $0.00002 $0.00249

Measured 17d ago against content hash 96b096192dd6, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

finishing-a-development-branch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 17d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

100% identical to finishing-a-development-branch — 221 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

scripts/skill-overlays/finishing-a-development-branch/SKILL.md · 25 lines

How it starts

The opening of the file, as written. The whole thing — 25 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Finishing a Development Branch

Confirm the final change has completed the project's independent quality-check and required checks. Evidence must describe the current tree: rerun affected verification after material edits or integration changes, and report unresolved failures. Do not add another whole-branch review or repeat an unchanged successful test suite merely to follow this skill.

Determine the repository, branch, base, worktree ownership, and uncommitted changes with read-only Git commands. Verify the intended target from the plan, upstream, or conversation before integrating.

Follow the integration action already authorized by the user. The design approval already covers commits, feature-branch pushes, creating the PR and the merge. First, if the project's rules require the user's approval for the merge, or the merge automatically releases, publishes or deploys to production - an environment that customers or users rely on (exception X2; an automatic deploy to an environment for checking behavior (such as staging) that does not touch production - the production database, live payments, or sending to real customers and the like - is not X2; judge by the effect on customers or users, not by the name) - finish the reviewable result and put that question in the final message. Only an answer to that item, or an instruction that names the operation and its target (range), counts as an X2 OK - not the design approval, a generic "continue", or an instruction that names no range. Otherwise look at the deviations from the design - both the spec / plan record and the PR body's deviation section. If there are none, merge with the project's merge procedure once quality-check is complete. If there are any, they change what was approved: finish everything else but create the quality-check flag only after the user's confirmation, ask in the final message as a separate decision ("merge with these deviations?"), and merge after the user's confirmation. A deviation that appears after that confirmation needs a new confirmation. Do not present an approval menu again. End with the final message of documents/development/development-policy.md §1.0 "承認後の進め方".

For a project that promotes main to a production branch (release or similar), the X2 question covers everything that goes out (§1.0 "承認後の進め方" 8). Before asking, run git fetch origin and fix the end SHA of the source. List the PRs not yet promoted up to that SHA: with merge-commit promotion, git log --first-parent --oneline origin/<release>..<SHA>; with squash promotion, main's commits never become ancestors of the production branch, so start from the previous source (the SHA recorded in the last promotion PR, or the first-parent main commit whose tree equals origin/<release>^{tree}); if no starting point is found, do not guess a list - write that it was not found in the final message. Show the list and the short SHA in the final message. As for any X2 OK, only an answer to the item that shows the list and the SHA, or an instruction that names the operation and the range, counts (on a "promote it" that names no range, build the list and the SHA and ask). Promote exactly that SHA; anything that lands before or after the OK stays out. When the promotion goes through a PR, do not use main as its head (push a temporary branch at the SHA and use it, for example git push origin <SHA>:refs/heads/promote/<short SHA>), or merge with gh pr merge --match-head-commit <SHA>. With squash promotion, write the source SHA in the promotion PR body (the next starting point). When this PR is not merged yet, add "this PR (after merge)" to the list; after the merge, if the first-parent range <the SHA you asked with>..<this PR's merge commit> holds only this PR, use that merge commit as the end, otherwise rebuild the list and ask again. After the OK, the AI merges and promotes; do not hand those operations back. The commands depend on the merge method: the project's own procedure holds the exact steps.

Read the full file on GitHub · 25 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 17d ago Changed · +4 lines 96b096192dd6
  2. 18d ago Changed 80accf9a7b4b
  3. 20d ago Changed · +2 lines 65cdf0d01f59
  4. 1mo ago Changed · -207 lines · -7 tokens per session 87d84ae45ddf
  5. 1mo ago First seen · 226 lines · 27 tokens per session scan A 8db5a922b242

Subscribe to this mod's changes

finishing-a-development-branch is a skill published in the GitHub repository Crearize/ai-dev-helm (5 stars, last pushed today), licensed MIT. It adds 20 tokens to every session and 2,490 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 100% identical to finishing-a-development-branch, differing in 221 lines, and is treated as a copy.