repo-health
Skill1 starsChanged 37 days agoMITView on GitHub
Check this repo for the drift that hides real problems — unlanded branches, uncommitted work piling up, red CI on main, placeholder modules the docs describe as delivered, stale planning docs, and dead routes. Use when asked for a status report, a health check, "where are we", "what's half built", "anything drifting", or before landing a branch, cutting a release, or writing a status update. Also use at the start of a session that will touch this repo broadly.
6 other projects use a version of this.
Adopted by
Most recently taken up by
- dataGriff/skillstheir own version, first seen Nov 1, 2026
- damsleth/SKILLStheir own version, first seen Oct 25, 2026
- moberghr/mtk-agent-toolkittheir own version, first seen Oct 25, 2026
- Sassy-Dog/skillstheir own version, first seen Oct 8, 2026
- exerias21/brainstorm-toolkittheir own version, first seen Oct 4, 2026
- calebnewtonusc/Chewbaccatheir own version, first seen Aug 11, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- calebnewtonusc/Chewbacca★ 8same name
- moberghr/mtk-agent-toolkit★ 8same name
- damsleth/SKILLS★ 1same name
- Sassy-Dog/skills★ 1same name
- dataGriff/skills★ 0same name
- exerias21/brainstorm-toolkit★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 61
Update STATUS.md — the living record of where this project actually is.
- Distinctive: line 66
If a check fires because the *threshold* is wrong rather than the repo, change the threshold at the top of the script and say why in the …
- Distinctive: line 68
Do not delete a check to make the output green.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Repo health
On 9 September 2026 this repo was audited and found in a state no single
commit caused: main had not moved in 51 days, 88 commits and ~4,900
uncommitted lines sat on one branch, docs/ROADMAP.md described a plan
that had been abandoned six weeks earlier without anyone writing that
down, CI on main had been red since July, and docs/VISION.md called a
pipeline "the non-negotiable promise" while four of its five stages were
six-line placeholder functions.
None of that was a bug. Every individual commit was fine. The failure was that nothing measured the gap between what the repo said and what it was, so the gap grew for two months unnoticed.
This skill is that measurement.
Run it
scripts/repo-health.sh # fast, no network, ~2s
scripts/repo-health.sh --full # + cargo check, clippy -D warnings, frontend buildExit codes: 0 clean · 1 warnings · 2 breach. The --full gates are the
same ones CI runs, so a green --full means a green pipeline.
What it checks, and why each one is here
| # | Check | The failure it would have caught |
| 1 | Unlanded work | main frozen 51 days; a 88-commit branch nobody could review |
| 2 | Uncommitted volume | 4,921 lines and two whole modules living only in one working tree |
| 3 | CI on main | red since July, invisible because nobody looked |
| 4 | Stub drift | four placeholder modules shipping while docs called them delivered |
| 5 | Doc drift | a roadmap 46 days behind the code, describing an abandoned plan |
| 6 | Dead surfaces | /data saying "Coming soon", /api/compat/* with no caller |
Reading the output
BREACH means stop and deal with it before continuing the current task. The two that matter most:
- main has not moved in N days — land what is landable today. Work that never lands exists on exactly one disk.
- doc is N days behind the code — this is the dangerous one, because it is silent. A plan nobody follows is worse than no plan: it hides that a decision was made. Fix it by editing the doc to describe what actually happened, not by deleting the check.
WARN means look at it this week. Placeholder modules and dead routes are not urgent, but each one is a promise the product does not keep, and they only ever accumulate.
After running it
Update STATUS.md — the living record of where this project actually is.
It exists so the next person (or the next session) does not have to
re-derive the state from 99k lines of Rust. Keep it honest about what is
stubbed; that honesty is the whole point.
If a check fires because the threshold is wrong rather than the repo, change the threshold at the top of the script and say why in the commit. Do not delete a check to make the output green.
When not to use this
It is a repo-level smoke alarm, not a code reviewer. For defects in a
specific change, use /code-review. For a design question about one
subsystem, read the module's header comment — they are unusually good in
this codebase and are the intended entry point.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

