Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add bacchus-labs/wrangler/plugin install wranglerWrote 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.
[](https://agentmods.dev/skills/bacchus-labs/wrangler/isolating-worktrees)<a href="https://agentmods.dev/skills/bacchus-labs/wrangler/isolating-worktrees"><img src="https://agentmods.dev/badge/skills/bacchus-labs/wrangler/isolating-worktrees.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00035 | $0.02252 |
| Opus 5 | $0.00017 | $0.01126 |
| Sonnet 5 | $0.00007 | $0.00450 |
| Haiku 4.5 | $0.00003 | $0.00225 |
Grade A, and why
isolating-worktrees 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 347 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Worktree Isolation
Overview
Git worktrees provide isolated workspaces, but Claude Code has characteristics that can cause changes to "bleed" to the wrong worktree:
- cwd resets between Bash calls - Each Bash tool invocation starts in the original working directory
- Subagents don't inherit verified context - They start with zero knowledge of parent's location
- No automatic verification - Nothing enforces you're in the right place before git operations
This skill provides protocols to ensure ALL changes stay in the intended worktree.
The Problem
Why Changes Bleed
# WRONG - This DOES NOT work in Claude Code
cd /path/to/worktree
git status # This runs in ORIGINAL directory, not worktree!
Each Bash tool call is independent. The cd command affects only that shell session, which ends immediately. The next Bash call starts fresh in the original working directory.
Subagent Context Loss
When you spawn a subagent with:
Work from: /path/to/worktree
This is just text. The subagent has no verified context about which directory it should use. If it runs git commit, it commits wherever Claude Code's process is running - likely your main branch.
The Solution: Verification Protocol
Rule 1: Always Chain Commands
WRONG:
cd /path/to/worktree
git add .
git commit -m "message"
RIGHT:
cd /path/to/worktree && git add . && git commit -m "message"
Or use git's -C flag:
git -C /path/to/worktree add .
git -C /path/to/worktree commit -m "message"
Rule 2: Verify Before Every Git Operation
Before ANY git write operation (add, commit, push, merge, etc.), run verification:
cd /path/to/worktree && \
echo "=== WORKTREE VERIFICATION ===" && \
echo "Current directory: $(pwd)" && \
echo "Git root: $(git rev-parse --show-toplevel)" && \
echo "Branch: $(git branch --show-current)" && \
echo "Expected: /path/to/worktree on branch-name" && \
echo "============================"
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.
- 6d ago First seen · 347 lines · 35 tokens per session scan A d7b6dbbc61be
isolating-worktrees is a skill published in the GitHub repository bacchus-labs/wrangler (4 stars, last pushed 6mo ago), licensed MIT. It adds 35 tokens to every session and 2,252 once invoked, about $0.0002 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.
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.
gh-auth-isolation
Safely manage multiple GitHub identities (EMU + personal) in agent workflows.
comet-github
将 Comet GitHub 维护请求路由到基于证据的 PR 审阅、Issue 分诊、本地想法收集、CI 诊断或 Issue 实施流程。用户提到 Comet GitHub Issue/PR 但未指定流程,或询问下一步如何处理时使用。.
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.
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…
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".