knowledge-refresh
Skill84 starsChanged 13 days agoMITView on GitHub
Use after a release, to compare the knowledge Claude Code reads, CLAUDE.md, the standards, the source pages and the incident log, with the code and the data, and propose what is stale, missing or wrong.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- dmzoneill/skillstheir own version, first seen Oct 24, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- dmzoneill/skills★ 2same 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
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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 23
Keep it short: what repeats another file is cut and linked.
- Distinctive: line 40
Never applies a change without the yes.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Knowledge refresh
The job: keep the knowledge that Claude Code reads true to the project: CLAUDE.md and the knowledge folder. It proposes, and after the data engineer's yes it writes the approved changes. The runbook and the how-to guides belong to the docs skill.
Input: none, or the area to check.
Steps
- CLAUDE.md: every folder it names exists, every skill is listed, every path is right.
- The standards against the code: naming, layers, code style.
- The sources against the contracts and the files: one page per source, current.
- The business: the glossary and the stakeholders against the documents.
- The architecture and the core model against what is built.
- The known issues and the incident log: what a run has proven since. The incident skill adds the entries, this skill checks them.
- The decision log: what is decided, what is still open.
- A document nobody has read yet: what it adds to the knowledge.
- Keep it short: what repeats another file is cut and linked.
- Report every finding as a proposal: the file, what it says, what is true.
Output
A list of proposals, one per finding: the file, what it says, what is true, and the change. After the yes, the approved changes in the knowledge files.
Reads
- the knowledge, the code, the data contracts, the documents, the run records
Writes
- nothing until the yes, then only the knowledge files the data engineer approved
Never
- Never applies a change without the yes.
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.