edit
Skill1.7k starsChanged 34 days agoMITView on GitHub
Add or remove raw sources, or update wiki content, per user request
We split this skill into the rules it gives an agent
5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 10 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 32
If the user provides a local path: copy to the corresponding directory under raw/
- Very specific: line 33
If the user provides an arXiv URL: download to raw/papers/
- Very specific: line 34
If the user provides a web URL: fetch content with markdownify and save to raw/web/
- Distinctive: line 38
Read the relevant pages and modify content per user instructions
- Distinctive: line 44
When writing forward links, simultaneously write reverse links
“Common” is measured: kept by at least 6 in 10 of the 10 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
/edit
Add or remove raw sources, or update wiki content, per user request.
Trigger
User manual: /edit <user request>
Inputs
User request, for example:
- "Download this paper to raw/papers/"
- "Delete raw/papers/xxx.pdf"
- "Update the SOTA tracker in topics/efficient-llm-adaptation"
- "Add a new variant to concepts/lora"
Outputs
Updated wiki files, index.md, log.md
Steps
STEP 1: Parse User Intent
- Add raw sources:
- If the user provides a local path: copy to the corresponding directory under
raw/ - If the user provides an arXiv URL: download to
raw/papers/ - If the user provides a web URL: fetch content with markdownify and save to
raw/web/
- If the user provides a local path: copy to the corresponding directory under
- Delete raw sources:
- Confirm then execute deletion
- Update wiki:
- Read the relevant pages and modify content per user instructions
STEP 2: Execute Updates
- Newly added raw sources can later be incorporated into the wiki via
/ingest - Direct wiki modifications: update the specified fields/content in specific pages per user instructions
- When writing forward links, simultaneously write reverse links
STEP 3: Update Navigation
EDIT wiki/index.md: update relevant entriesAPPEND wiki/log.md:## [{date}] update | {description}
STEP 4: Report
- List all changes made
- Suggest follow-up actions (e.g. ingest newly added raw sources if applicable)
Constraints
raw/is read-only for existing files (this skill may add files toraw/, but must not modify existing ones)- Wiki modifications must follow template structure
- Bidirectional links must be kept in sync
Trending
More talk lately.thin dataabout the whole repository
Who's talking about PKU-DAIR/AutoSci
The biggest 1 of 1 mention linking to the repository that our crawl found. Each line is their own headline.
Other versions15 projects use a version of this.
Word for word · 3
- DAIR-AutoResearch/AutoSci★ 1.7k100% same
- skyllwt/AutoSci★ 1.7k100% same
- skyllwt/OmegaWiki★ 1.7k100% same
Edited · 0
None found.
Wrote their own version · 12
- PrimeIntellect-ai/prime-agent★ 22ksame name
- Lambenthan/empiricalwiki★ 85same name
- KingGyuSuh/codex-image-in-cc★ 35same name
- howells/fiction★ 19same name
- Erickuby/Social-Media-Autoposting-With-Claude-Code★ 17same name
- Agents4Academia-AI/UReKA★ 9same name
- gagoar/iceberg★ 8same name
- justinseger015-ctrl/ix-ai-agent-social-media-manager★ 2same name
- EdwardAF-IT/claude-skills★ 1same name
- Jakeschincariol/master-skills★ 1same name
2 more not listed
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 repository30 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- runtime/CLAUDE.mdCLAUDE.md
- askSkill
- checkSkill
- daily-arxivSkill
- discoverSkill
- exp-designSkill
- exp-evalSkill
- exp-pilot-evalSkill
- exp-pilot-runSkill
- exp-runSkill
- exp-statusSkill
- hiera-experimentSkill
- ideateSkill
- ingestSkill
- initSkill
- noveltySkill
- paper-compileSkill
- paper-draftSkill
- paper-planSkill
- posterSkill
- prefillSkill
- rebuttalSkill
- refineSkill
- researchSkill
- resetSkill
- reviewSkill
- setupSkill
- surveySkill
- visualizeSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.