thrashr888/shift/checkout-health
Claude Code and agent skill by thrashr888/shift
- community
Checking that the SKILL.md file is still in the repository.
Check a Git checkout's health by reporting whether the working tree is clean and what the recent commits are; use when asked to assess, summarize, or sanity-check the current repo state before or after work.
Install
npx vskill@latest install thrashr888/shift/checkout-healthSKILL.md
checkout-health
Report a Git checkout's health in one or two sentences: whether the working tree is clean (no uncommitted changes) and what the last commit was.
When to use
- User asks to "check the checkout", "is the tree clean", "what did we last commit",
or to summarize repo state before/after a change.
- Pre-flight before a risky task or before reporting work done.
- Triage: distinguish a dirty tree (needs attention) from a clean one.
Steps
- Run
git status --shortandgit log --oneline -3in parallel (independent reads).
In this runtime, use the run tool with an argv list: ["git","status","--short"] and ["git","log","--oneline","-3"].
- Read the status output:
- No output lines → tree is clean.
- Any line (
M,A,D,??,R, etc.) → tree is dirty; list the paths.
- Read the log output: the first line is the last commit. Take its short SHA and subject.
- Summarize in two sentences:
- Sentence 1: clean vs. dirty, naming the dirty paths (if any).
- Sentence 2: "The last commit was
<sha>—<subject>."
Commands that worked
Both returned exit 0 and are safe read-only commands.
Pitfalls
- Not clean ≠ broken. A dirty tree just means uncommitted changes; report it, don't fix.
- Read the paths, don't guess. Quote the exact paths from
git status --short. - Bare SHA + subject.
--onelinegives<sha> <subject>; use that, not full hashes. - Not a repo / no commits. If a command exits non-zero, report the error instead of a
clean/dirty verdict. See reference.md for edge cases.
- argv, not shell. Prefer the
runtool with an argv list so--and paths are
not mangled by shell parsing.
What to verify
- Confirm exit code 0 for both commands (or report the real error).
- Confirm the two-sentence summary matches: status output → clean/dirty verdict, and
first log line → last commit SHA + subject.
- No secrets or machine-specific paths in the report.
See reference.md for the full status-code table, edge cases, and examples.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryChecking.shift/skills/checkout-health/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/thrashr888/shift/checkout-health)