Back to search
skillzbench
MarketplaceWorks with Claude CodeGitHub
SkillzBench for Claude Code: benchmarked, security-scanned agent skills through one MCP connector.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/skillzbench--claude-plugin)HTML
<a href="/listings/skillzbench--claude-plugin"><img src="/api/listings/skillzbench--claude-plugin/badge.svg" alt="findsafeskills safety rating" /></a>Install
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
.claude-plugin/marketplace.json · Manifest · 21 lines
- {
- "name": "skillzbench",
- "owner": {
- "name": "SkillzBench (AIQ Labs)",
- "email": "team@skillzbench.com"
- },
- "metadata": {
- "description": "Benchmarked, security-scanned agent skills for Claude Code, served through one MCP connector."
- },
- "plugins": [
- {
- "name": "skillzbench",
- "source": "./plugins/skillzbench",
- "description": "Connect Claude Code to SkillzBench: 1,000+ security-scanned agent skills, benchmarked with vs. without the skill. Switch skills on at skillzbench.com and Claude loads the right one when a task needs it.",
- "version": "1.0.0",
- "homepage": "https://skillzbench.com/agents",
- "category": "productivity",
- "keywords": ["skills", "agent-skills", "mcp", "benchmarks", "security"]
- }
- ]
- }
README.md · README · 25 lines
- # SkillzBench for Claude Code
- Benchmarked, security-scanned agent skills for Claude Code, served through one MCP connector.
- ```text
- /plugin marketplace add SkillzBench/claude-plugin
- /plugin install skillzbench@skillzbench
- ```
- Then run `/mcp`, pick **skillzbench**, choose **Authenticate**, and sign in (free). Pick skills or a starter pack at
- [skillzbench.com](https://skillzbench.com/skills). Claude sees a short list of what you switched on and loads a skill
- only when a task calls for it.
- **No account yet?** Try the read-only starter set first:
- ```bash
- claude mcp add --transport http skillzbench-trial https://mcp.skillzbench.com/try
- ```
- - Every skill is security-scanned and pinned to the scanned version.
- - 236 skills are benchmarked head-to-head, with vs. without the skill; wins and losses are both published.
- - Skills that bundle scripts are served only when the license is permissive and the scan is clean. Claude shows you a
- script before running it.
- [How we measure](https://skillzbench.com/methodology) · [All skills](https://skillzbench.com/skills) · team@skillzbench.com
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.