update-feature
Skill1 starsChanged 3 months agoMITView on GitHub
Update an existing feature while keeping its tests green. Use when the user asks to change, extend, modify, or adjust behaviour that already exists — phrasings like "update the X", "make X also do Y", "change how X works", "add Y to X", "tweak X", "X should also handle Y", "make X do Y instead", "extend X to cover Y", "adjust X so that Y", "rework X", "X needs to also do Y", "can you make X do Y", "while keeping the tests green". Use it for any request to change what existing code does, rather than to write something new. Not for renames, formatting or docs-only edits.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- quocnt59/claude_skilltheir own version, first seen Aug 6, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- quocnt59/claude_skill★ 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.
In this repository6 other files it gives its agents.
- build-newSkill
- debug-failureSkill
- explain-codeSkill
- plan-approachSkill
- review-diffSkill
- write-testsSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Update this feature: $ARGUMENTS.
- Distinctive: line 11
Keep the tests for behaviour you are not changing passing.
- Distinctive: line 22
Run the feature's tests before you change anything, and the whole suite where that is quick, and note what already fails.
- Distinctive: line 26
If the change adds behaviour that isn't covered, add a test for it alongside.
- Distinctive: line 27
Run the feature's tests again, then the whole suite, on the final code and tests.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Update a feature
Update this feature: $ARGUMENTS. If that is empty, work from the user's last message.
Keep the tests for behaviour you are not changing passing. If this turns out to be a rename, a formatting change or a docs-only edit, just make it: the steps below are for changes in behaviour.
How to work
- Find the feature and read it before changing anything. Read its tests too — they are the specification you must not break. If it has no tests, say so, and first pin the behaviour you are about to rely on with a test or two (with no test setup at all, record a before-and-after run on a fixture or a dry run), so the change is checked against something. If it cannot be run safely, say so instead of running it.
- Run the feature's tests before you change anything, and the whole suite where that is quick, and note what already fails. Those are not yours to fix here: report them.
- Make the smallest change that delivers the request. Match the surrounding code's naming, comment density, and idiom.
- If the change adds behaviour that isn't covered, add a test for it alongside.
- Run the feature's tests again, then the whole suite, on the final code and tests.
Fix the failures you caused; for a failure outside the feature's tests, work
out whether it fails without your change before treating it as yours. Read the diff
first. If you have to take your edit out to check, do it by hand (never
git checkout,git restoreorgit stash, which throw away uncommitted work), put it back, confirm the file is your finished version again, and rerun the final checks. If you cannot tell, report the failure as unattributed. Do not edit a test to make it pass unless the request itself changes what that test exercises; then change only the setup, calls and assertions the request makes necessary. Never skip a test, and delete one only when the request removes the behaviour it asserts.
Report back
State what changed, the exact test command and its summary line (passed, failed, skipped), and every test you edited or deleted, with its old and new expectation. List failures that were there before you started separately. If tests failed and you could not fix them, or could not be run, say so with the output rather than reporting success.
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.

