Skill Curator
Pricing
from $10.00 / 1,000 results
Skill Curator
Search 117,000+ Claude Code Skills and get quality-scored recommendations. Each result is enriched with live GitHub metadata and scored across 4 dimensions: popularity, freshness, documentation quality, and license compliance. Pay only for what you use — $0.05 per search + $0.01 per result.
Pricing
from $10.00 / 1,000 results
Rating
5.0
(1)
Developer

AI Crew Solutions
Actor stats
0
Bookmarked
9
Total users
6
Monthly active users
14 days ago
Last modified
Categories
Share
Skill Curator - Find the Best Claude Code Skills
Search 117,000+ Claude Code Skills and get quality-scored recommendations with GitHub metadata enrichment. Stop scrolling through thousands of skills manually — describe what you need in plain English and get ranked results in seconds.
What does Skill Curator do?
Skill Curator searches the SkillsMP database of 117,000+ Claude Code Skills, then enriches each result with live GitHub data to calculate a quality score. Unlike basic keyword search, Skill Curator evaluates every skill across 4 dimensions:
| Quality Dimension | Weight | What it measures |
|---|---|---|
| Popularity | 25% | GitHub stars (log-scale normalized) |
| Freshness | 25% | Days since last push to repository |
| Documentation | 35% | SKILL.md presence, structure, and completeness |
| License | 15% | Open-source license availability |
Each result includes a 0-100 quality score so you can instantly identify the best skills for your workflow.
Why use Skill Curator?
Claude Code Skills supercharge your AI coding workflow, but finding the right ones from 117,000+ options is overwhelming. Skill Curator solves this by:
- Saving hours of research — One search replaces manual browsing across GitHub and SkillsMP
- Quality verification — Every result is checked against live GitHub data (not stale metadata)
- Smart ranking — Results sorted by a multi-factor quality score, not just star count
- AI-powered search — Optional semantic search understands natural language queries like "help me write unit tests" or "create PDF reports"
Use cases
- Developers setting up Claude Code — Find the best skills for your tech stack
- Teams standardizing workflows — Identify well-maintained, documented skills for team adoption
- Skill authors — Research existing skills before building your own
- Automation pipelines — Integrate skill discovery into your toolchain via Apify API
How to search for Claude Code Skills
- Enter your search query describing what you need (e.g., "PDF creation", "React testing", "database management")
- Set the maximum number of results (1-20, default 10)
- Optionally enable AI Semantic Search for better relevance on complex queries
- Click Start and get enriched, quality-scored results in ~6 seconds
Input parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
| Search Query | string | (required) | Natural language description of skills you need |
| Max Results | integer | 10 | Number of enriched results to return (1-20) |
| Include AI Search | boolean | false | Enable semantic search for better relevance (+~8s) |
| Minimum Stars | integer | 0 | Filter out skills below this GitHub star count |
Input example
{"query": "PDF creation and document management","limit": 5,"includeAiSearch": false,"minStars": 10}
Output
Each result contains the original skill data plus live GitHub enrichment and a calculated quality score.
Output fields
| Field | Description |
|---|---|
name | Skill name |
author | Skill author |
description | Short description of what the skill does |
githubUrl | Link to the skill's GitHub repository |
skillUrl | Link to the skill's SkillsMP page |
qualityScore.overall | 0-100 composite quality score |
qualityScore.popularity | Popularity sub-score (based on GitHub stars) |
qualityScore.freshness | Freshness sub-score (based on last push date) |
qualityScore.documentation | Documentation sub-score (based on SKILL.md analysis) |
qualityScore.license | License sub-score (open-source check) |
repoMetadata.stars | Current GitHub star count |
repoMetadata.forks | Fork count |
repoMetadata.pushedAt | Last push date |
repoMetadata.archived | Whether the repo is archived |
repoMetadata.license | SPDX license identifier |
skillMdAnalysis | SKILL.md structure analysis (front matter, tools, examples) |
enrichedAt | ISO timestamp of when enrichment was performed |
Output example
{"name": "working-with-documents","author": "anthropics","description": "Suite of tools for creating elaborate, multi-component HTML artifacts","githubUrl": "https://github.com/anthropics/courses/tree/master/claude-code-skills/07-working-with-documents","skillUrl": "https://skillsmp.com/skills/anthropics/working-with-documents","repoMetadata": {"stars": 8234,"forks": 1205,"openIssues": 42,"license": "MIT","updatedAt": "2026-01-28T10:30:00Z","pushedAt": "2026-01-25T14:22:00Z","archived": false,"description": "Anthropic's educational courses"},"skillMdAnalysis": {"hasYamlFrontMatter": true,"hasDescription": true,"hasAllowedTools": true,"contentLength": 4520},"qualityScore": {"overall": 65,"popularity": 80,"freshness": 70,"documentation": 55,"license": 100},"enrichedAt": "2026-01-31T12:00:00Z"}
How much does it cost?
Skill Curator uses a Pay Per Event pricing model — you only pay for what you use.
| Event | Price | Description |
|---|---|---|
| Actor Start | $0.05 | Charged once per search |
| Each Result | $0.01 | Charged per enriched skill returned |
Cost examples
| Scenario | Results | Cost |
|---|---|---|
| Quick search (5 results) | 5 | $0.10 |
| Standard search (10 results) | 10 | $0.15 |
| Deep search (20 results) | 20 | $0.25 |
| Monthly usage (10 searches) | 100 | $1.50 |
Apify Free plan includes $5 of monthly usage credits — enough for 30+ searches at no cost.
Tips for getting better results
- Be specific: "React component testing with Jest" works better than "testing"
- Use English: The underlying database indexes English content most effectively
- Enable AI Search for natural language queries like "help me build a landing page with a contact form"
- Set Minimum Stars to filter out experimental or abandoned skills
- Start with default limit (10) — increase only if you need broader coverage
Integrations
Skill Curator outputs standard JSON to the Apify Dataset, which you can:
- Export as JSON, CSV, or Excel from the Apify Console
- Access via API at
https://api.apify.com/v2/datasets/{datasetId}/items - Connect to Make, Zapier, or n8n via Apify integrations
- Schedule recurring searches to monitor new skills in your domain
FAQ
What are Claude Code Skills?
Claude Code Skills are reusable prompt configurations that extend Claude Code's capabilities. They can add specialized knowledge, workflows, and tool integrations — from generating PDFs to managing databases to creating frontend designs.
Where does the data come from?
Skill Curator searches SkillsMP, the largest directory of Claude Code Skills (117,000+), then enriches results with live data from the GitHub API. The quality score is calculated in real-time based on current repository metadata.
How is the quality score calculated?
The score (0-100) is a weighted average of four factors: Popularity (25%, log-scale GitHub stars), Freshness (25%, days since last push), Documentation (35%, SKILL.md analysis), and License (15%, open-source availability). Higher scores indicate better-maintained, well-documented skills.
Can I use this in my automation pipeline?
Yes. Skill Curator exposes a standard Apify API. You can trigger runs programmatically, poll for results, and integrate with any system that speaks HTTP/JSON. See Apify API documentation for details.
What's the difference between keyword and AI search?
Keyword search (~1s) matches exact terms against skill names and descriptions. AI Semantic Search (~8s) uses vector embeddings to understand intent — useful for queries like "help me with database migrations" where exact keyword matches may miss relevant results.