wip

wip is a skill for Claude Code from motlin/claude-code-plugins. It costs 134 tokens per session (1,635 once invoked), scanned A, original, Apache-2.0.

A repository cleanup workflow for unfinished work spread across Git worktrees and branches. A worktree is another working directory linked to the same repository.

In plain words
What is it for?
Use it to remove stale worktrees, rebase branches, delete branches that are already merged, rerun tests after rebasing, and push branches that pass.
Why use it?
It brings leftover work toward a clean state while protecting uncommitted changes and unexpected remote updates.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /Users/you/projects/some-permanent-worktree.

Part of the git plugin — 12 skills shipped together

Good fit Use it to remove stale worktrees, rebase branches, delete branches that are already merged, rerun tests after rebasing, and push branches that pass.

Compare 6 skills from other repositories ↓
Install

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.

Claude Code
/plugin marketplace add motlin/claude-code-plugins
Claude Code
/plugin install git

Made for: Claude Code.

Or install git, the plugin that ships this one along with the rest of its 12 skills.

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 wip

README.md
[![agentmods](https://agentmods.dev/badge/skills/motlin/claude-code-plugins/wip/github.svg)](https://agentmods.dev/skills/motlin/claude-code-plugins/wip)
Your own site
<a href="https://agentmods.dev/skills/motlin/claude-code-plugins/wip"><img src="https://agentmods.dev/badge/skills/motlin/claude-code-plugins/wip/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 wip

Your own site · 80×15
<a href="https://agentmods.dev/skills/motlin/claude-code-plugins/wip"><img src="https://agentmods.dev/badge/skills/motlin/claude-code-plugins/wip.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 134 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,635 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 $0.00134 $0.01635
Opus 5.5 $0.00054 $0.00654
Sonnet 5.5 $0.00027 $0.00327
Haiku 4.5 $0.00013 $0.00163

Measured 9d ago against content hash 6f6ec10ad5d2, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

wip 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 9d 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.

plugins/git/skills/wip/SKILL.md · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/wip — work-in-progress cleanup pipeline

Drive a repo full of leftover worktrees and half-finished branches toward a clean, landed state. Four phases run in order, each gated:

  • Worktrees: remove every worktree except the main and protected ones, freeing their branches.
  • Rebase + GC: rebase every branch onto upstream main (j g / git-all) and delete merged branches.
  • Test: retest the rebased commits.
  • Push: push the green branches. The user opens and merges PRs.

Opening PRs, CI, review, and merging are the user's call. If the repo or user documents a WIP state model, follow it for those stages.

Golden Rules

  • Never --force. Remove worktrees with git worktree remove --no-force (write the flag explicitly) and push with a pinned --force-with-lease. Git's refusal is the safety net for uncommitted work and unexpected remote changes.
  • The worktree and branch set is a moving target. Other agents and the user create, switch, and merge while you work. Recompute with git worktree list --porcelain / git for-each-ref immediately before each action, and re-git fetch before testing and pushing.
  • Gate irreversible and outward-facing steps. Removing a worktree with WIP, force-pushing, and delete-merged (which deletes remote branches) need the user's OK. Ask at each phase boundary unless the user pre-authorized the whole flow.
  • Keep the main checkout clean. Rebases and tests check out commits, so run them in throwaway worktrees (git worktree add) and tear those down afterward.

Protected Worktrees

The main worktree is the first entry of git worktree list --porcelain (the one whose gitdir is git rev-parse --git-common-dir). Never remove it.

Permanent worktrees are listed per repo in .llm/wip.json:

{"protectedWorktrees": ["/Users/you/projects/some-permanent-worktree"]}

If the file is missing, only the main worktree is protected, but ask before removing any long-lived worktree you're unsure about. git worktree lock also works, since a locked worktree can't be removed without --force.

Read the full file on GitHub · 83 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. 9d ago Changed 6f6ec10ad5d2
  2. 22d ago Changed · -60 lines 42efb7034459
  3. 1mo ago Changed bdddf7c581ab
  4. 1mo ago First seen · 143 lines · 134 tokens per session scan A 92a2e018fe99

Subscribe to this mod's changes

wip is a skill published in the GitHub repository motlin/claude-code-plugins (20 stars, last pushed yesterday), licensed Apache-2.0. It adds 134 tokens to every session and 1,635 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-08-30.