skills / rules
Cursor rule1 starsChanged 48 days agoApache-2.0View on GitHub
Standard structure and writing conventions for skill documentation
We split this skill-writing-standard.mdc into the rules it gives an agent
7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 9
Use this structure for every skill README.md and SKILL.md.
- Very specific: line 13
Start each SKILL.md with valid YAML frontmatter containing name and a dense, selection-oriented description.
- Very specific: line 14
Use exactly one clear # title.
- Distinctive: line 15
Follow the title with a 2–4 sentence summary explaining what the subject is and when to use the document.
- Very specific: line 16
Add a short ## Quick reference section containing the facts needed most often.
- Distinctive: line 18
Put structured reference data in tables.
- Distinctive: line 19
Put examples in fenced code blocks.
≈ “Common” is an estimate: at least 1 in 5 of the 6,350 Cursor rule 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
---
description: Standard structure and writing conventions for skill documentation
globs: "**/{README.md,SKILL.md}"
alwaysApply: false
---
# Skill writing standard
Use this structure for every skill `README.md` and `SKILL.md`.
## Required structure
1. Start each `SKILL.md` with valid YAML frontmatter containing `name` and a dense, selection-oriented `description`. Keep frontmatter metadata-like.
2. Use exactly one clear `#` title.
3. Follow the title with a 2–4 sentence summary explaining what the subject is and when to use the document.
4. Add a short `## Quick reference` section containing the facts needed most often.
5. Organize the body around tasks. Prefer the predictable headings `Concepts`, `File format`, `Workflow`, `Rules`, `Examples`, and `Troubleshooting`; use specific task headings beneath them when useful.
6. Put structured reference data in tables.
7. Put examples in fenced code blocks.
8. End with `## Related documentation`; do not crowd the introduction with related links.
## Rules
Rule: State critical instructions as declarative sentences prefixed with `Rule:`.
Rule: Preserve explicit redundancy around dangerous distinctions. Repeat an important constraint in the relevant workflow or troubleshooting section when that prevents mistakes.
Rule: Do not shorten a document at the expense of operational detail, assumptions, edge cases, or gotchas.
Rule: Keep heading vocabulary predictable, but choose task-oriented subheadings that describe what the reader is doing.
## Example
```markdown
Rule: `inventory.json5` is authoritative. Do not infer artifact locations from the default directory structure.
```
Where files are headingNew Cursor rule files have more rules about planning and tracking tasks, and fewer about code style.
Of the Cursor rule files we read, the 3,978 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository24 other files it gives its agents.
- .cursor/rules/no-client-names.mdcCursor rule
- create-engine9-pluginSkill
- e9-api-keySkill
- e9-cliSkill
- e9-custom-fieldsSkill
- e9-delegate-domainSkill
- e9-dev-tasksSkill
- e9-eqlSkill
- e9-exportSkill
- e9-funnelSkill
- e9-global-messageSkill
- e9-inputSkill
- e9-inventorySkill
- e9-mcpSkill
- e9-modelSkill
- e9-person-idSkill
- e9-person-remoteSkill
- e9-pii-freeSkill
- e9-pluginSkill
- e9-reportsSkill
- e9-source-codeSkill
- e9-tasks-apiSkill
- e9-timelineSkill
- e9-upgradeSkill
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.

