2.5k
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill skill-store- _meta.json1.3 KB
- SKILL.md2.7 KB
Overview
This skill is a smart installation advisor for ClawHub that finds and recommends the best skill for a given need. It searches archived skills, runs security checks and code-quality evaluations, and produces a ranked comparison with a clear recommendation. The tool is self-contained with no external dependencies beyond ClawHub and the local scanner.
How this skill works
It searches ClawHub archives for candidates matching your query, installs the top N into a temporary workspace, and runs an automated security scan using skill-shield. It then evaluates code quality and documentation, computes weighted scores (security 40%, quality 30%, relevance 30%), and generates a Markdown and JSON comparison report with a recommended skill. Non-recommended candidates are removed by default unless you request otherwise.
When to use it
- You need a safe, vetted skill to perform a specific task
- You want to compare multiple similar skills before installing
- You require a security-checked recommendation for production use
- You want a concise comparison report to share with a team
- You must audit archived skill versions before restoring or backing up
Best practices
- Limit the number of top candidates (--top) to control evaluation time
- Provide a focused search query to improve relevance scoring
- Supply a local skill-shield scanner path if auto-detection fails
- Use --output-dir to save human-readable and structured reports for audits
- Set --keep-all when you want to inspect removed candidates manually
Example use cases
- Find a reliable weather skill and get a security and quality comparison
- Compare image-generation skills and export a ranked report for stakeholders
- Audit a security-related skill before installing into a sensitive environment
- Evaluate archived versions of a skill to decide which release to restore
- Batch-evaluate multiple candidate skills during a migration or backup project
FAQ
You need the clawhub CLI configured and access to the skill-shield scan.py script; Python 3.8+ is required.
How are candidates scored and ranked?
Scores combine security (40%), quality (30%), and relevance (30%). Security uses the scanner results and permission checks; quality measures code size, documentation completeness, tests, and structure.