skill-crawler
Skill5 starsChanged 8 months agoMITView on GitHub
Convert crawled documentation into usable skills. Use after crawling with playwright-cli to generate SKILL.md files. For skill format details, see opencode-dev.
In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- charm-vhsSkill
- cloud-foundation-fabricSkill
- d2Skill
- duckdbSkill
- github-profile-architectSkill
- jqSkill
- opencode-devSkill
- playwright-cliSkill
- skill-registry-syncSkill
- tech-writerSkill
- terminal-ui-engineerSkill
- terraformSkill
- worktrunkSkill
- writingSkill
The file
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 10 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
Distinctive · estimate
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
- Distinctive: line 25
After crawling documentation with playwright-cli, this skill helps you:
- Very specific: line 29
Generate: - Create SKILL.md with proper format
- Distinctive: line 30
Validate: - Ensure quality and completeness
- Distinctive: line 36
Use playwright-cli to extract documentation:
- Distinctive: line 56
Create a new directory for your skill following the standard structure:
- Distinctive: line 130
Use "See also" sections
- Distinctive: line 134
Check extracted snapshots for hidden examples
- Distinctive: line 135
Look for "Getting Started" sections
- Distinctive: line 136
Search for "Example" headers
- Distinctive: line 141
Not by technical category (unless that helps)
- Distinctive: line 142
Test with a real task
≈ “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
Skill Crawler - Generate Skills from Documentation
Convert crawled documentation into reusable opencode skills.
Prerequisites
This skill works AFTER you have crawled documentation. See related skills:
- Crawl docs:
npx skills add https://github.com/benjaminwestern/google-engineer-skills --skill playwright-cli- Browser automation and content extraction - Skill format:
npx skills add https://github.com/benjaminwestern/google-engineer-skills --skill opencode-dev- YAML frontmatter, structure, and best practices
Overview
After crawling documentation with playwright-cli, this skill helps you:
- Analyze - Review extracted snapshots/content
- Structure - Organize into skill sections
- Generate - Create SKILL.md with proper format
- Validate - Ensure quality and completeness
Workflow
Step 1: Crawl documentation
Use playwright-cli to extract documentation:
playwright-cli open https://docs.example.com --persistent
playwright-cli snapshot --filename=overview.yaml
playwright-cli goto https://docs.example.com/commands
playwright-cli snapshot --filename=commands.yaml
playwright-cli closeStep 2: Analyze Extracted Content
Review your snapshot files and identify:
- Core commands - Main functionality
- Common workflows - Step-by-step processes
- Configuration - Setup and options
- Examples - Real usage patterns
Step 3: Create Skill Structure
Create a new directory for your skill following the standard structure:
mkdir -p skills/<skill-name>
touch skills/<skill-name>/SKILL.mdStep 4: Generate SKILL.md
Basic template:
---
name: <skill-name>
description: <1024 char description with "Use when...">
---
# <Service/Tool Name>
## Quick Start
```bash
<3-5 most common commands>Commands
Category 1
<commands>Examples
Example 1: <Common task>
<step-by-step>Common Workflows
<Description and steps>Tips
<Best practices>
```References
See the references/ folder for detailed guides:
- templates.md - Ready-to-use templates for CLI tools and API services
- best-practices.md - Writing guidelines, organization strategies, and do's/don'ts
- examples.md - Complete example skills demonstrating best practices
Validation Checklist
- YAML frontmatter is valid
- Name matches directory
- Description is under 1024 chars
- Quick start has working commands
- Examples are complete
- Categories are logical
- No markdown syntax errors
Troubleshooting
Content is too long
- Focus on most common 20% of commands
- Link to external docs for edge cases
- Use "See also" sections
Missing examples
- Check extracted snapshots for hidden examples
- Look for "Getting Started" sections
- Search for "Example" headers
Unclear organization
- Group by user intent (what they want to do)
- Not by technical category (unless that helps)
- Test with a real task
Related Skills
This skill is part of a documentation-to-skill workflow:
| Skill | Purpose |
npx skills add https://github.com/benjaminwestern/google-engineer-skills --skill playwright-cli | Crawl and extract documentation from websites |
npx skills add https://github.com/benjaminwestern/google-engineer-skills --skill opencode-dev | Skill format specs, agents, tools, and configuration |
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.

