worktree-discipline

worktree-discipline is a skill for Claude Code, Codex from onsager-ai/dev-skills. It costs 193 tokens per session (948 once invoked), scanned A, original, MIT.

A machine-wide Git safety rule that stops commits to the main branch and requires AI agents to work in Git worktrees, which are separate working directories for branches.

In plain words
What is it for?
Use it to install a single pre-commit guard for supported repositories and allow normal work on feature branches or linked worktrees.
Why use it?
It prevents accidental commits to shared or protected branches and keeps parallel agent changes isolated from the main checkout.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Claude Code.

Good fit Use it to install a single pre-commit guard for supported repositories and allow normal work on feature branches or linked worktrees.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/onsager-ai/dev-skills/worktree-discipline
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 onsager-ai/dev-skills --skill worktree-discipline
Clone the repo
git clone --depth 1 https://github.com/onsager-ai/dev-skills

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 worktree-discipline

README.md
[![agentmods](https://agentmods.dev/badge/skills/onsager-ai/dev-skills/worktree-discipline.svg)](https://agentmods.dev/skills/onsager-ai/dev-skills/worktree-discipline)
Your own site
<a href="https://agentmods.dev/skills/onsager-ai/dev-skills/worktree-discipline"><img src="https://agentmods.dev/badge/skills/onsager-ai/dev-skills/worktree-discipline.svg" alt="Measured on agentmods" height="20"></a>
Per session 193 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 948 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 original No closer match found 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 $0.00193 $0.00948
Opus 5 $0.00097 $0.00474
Sonnet 5 $0.00039 $0.00190
Haiku 4.5 $0.00019 $0.00095

Measured 7d ago against content hash e78a18dfa726, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

worktree-discipline 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 7d 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.

skills/worktree-discipline/SKILL.md · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

worktree-discipline

One global hook, every repo on the machine, no duplication. Companion to worktree-devproxy (which provides the wt new <branch> remedy the hook points at).

The policy

Who Where Result
anyone main / master blocked
agent (CLAUDECODE=1) any branch in the main checkout blocked — work in a worktree
agent linked worktree, feature branch allowed
human feature branch in the main checkout allowed
human ALLOW_MAIN_COMMIT=1 git commit or --no-verify escape hatch

Scope is automatic: the guard only fires in repos carrying a committed .envrc with export DEV_HOST= — the marker every worktree-devproxy onboarded repo already has. Other repos on the machine are untouched (the hook silently delegates), and every future onboarded repo is covered with no extra step.

Install (once per machine)

cp scripts/pre-commit ~/.githooks/pre-commit
chmod +x ~/.githooks/pre-commit
git config --global core.hooksPath ~/.githooks

Design notes (why it's shaped this way)

  • Global core.hooksPath, not per-repo hooks. One copy of the policy; updating it is editing one file. Per-repo .githooks copies drift and need a commit in every repo for every policy change.
  • Delegation keeps repo-specific hooks alive. Global hooksPath replaces .git/hooks resolution, so the global hook ends by exec-ing the repo's committed .githooks/pre-commit if present (e.g. onsager's migration-sequence check). Repos must NOT set a local core.hooksPath — local config overrides global and would silently drop the guard (watch for repo setup recipes that do this, e.g. a just setup running git config core.hooksPath .githooks).
  • Agent detection is CLAUDECODE=1 — set in every Claude Code Bash session. Other agent runtimes can be added to the same test.
  • Main-checkout detection: git rev-parse --git-dir equals --git-common-dir only in the primary checkout; in a linked worktree they differ.
  • The remedy is in the error message (use: wt new <branch>), so an agent hitting the wall self-corrects without reading docs.

Read the full file on GitHub · 66 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 7d ago First seen · 66 lines · 193 tokens per session scan A e78a18dfa726

Subscribe to this mod's changes

worktree-discipline is a skill published in the GitHub repository onsager-ai/dev-skills (5 stars, last pushed 25d ago), licensed MIT. It adds 193 tokens to every session and 948 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

prowler-commit

Creates professional git commits following conventional-commits format. Trigger: When creating commits, after completing code changes, when user asks to commit.

prowler-cloud/prowler · 33 tokens

gh-auth-isolation

Safely manage multiple GitHub identities (EMU + personal) in agent workflows.

github/gh-aw · 20 tokens

comet-github

A routing guide for Comet-related GitHub work. It directs requests about pull requests, issues, CI failures, ideas, and fixes to the appropriate review or implementation process.

rpamis/comet · 72 tokens

github-skill

Work with GitHub via the gh CLI — clone repositories, create/list/merge pull requests, create/list issues, and run any other gh command (API calls, workflow runs, releases, repo administration). List operations return parsed JSON.

zeenie-ai/OpenCompany · 51 tokens

codex-autoresearch

Run autonomous, measurable experiments in a Git repository: change one hypothesis, verify a numeric metric, keep improvements, and revert failures. Use when the user wants Codex to keep iterating toward a numeric target in the foreground or as a detached background run. Do not use for ordinary one-shot coding…

leo-lilinxiao/codex-autoresearch · 80 tokens

changelog-composer

Generates structured changelogs and release notes from git history and PRs, classifying breaking changes, features, fixes, performance, docs. Triggers on: "generate changelog", "write release notes", "what changed since", "prepare release", "release notes for", "diff since tag".

Mathews-Tom/armory · 67 tokens