mcpbeat Sign in

Claude Skills

The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.

Every Agent Skill we could find on GitHub, deduplicated by content. 86 696 files from 2 113 authors, of which 68 413 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.

68 413
unique skills
out of 86 696 files found on GitHub
18 283
are copies
same content, someone else's repository
1 744
tokens, median
what a typical skill costs you in context
8 409
name collisions
two skills with one name cannot sit side by side

68 401–68 413 of 68 413

page 1141 of 1 141
Sort Most copied
Most copied how often it was repackaged elsewhere Most stars on the repository that hosts it Lightest first cheapest to keep in context Heaviest first the ones that cost you the most Recently updated last push to the repository
Azure Resources Query
by microsoft
vendor

Use Azure Resources Query for read-only Azure resource lists: Show my Function Apps in Development; only those in West Europe; list Web Apps; find VMs; show storage accounts; browse resources in a subscription. Use explicit subscription scope and saved views. Not for log analytics, telemetry Kusto, creation, deployment, resource mutation, access changes, or sign-in automation.

1.4k tokens
Azure Functions Hosted Skills Canvas
by microsoft
vendor

MANDATORY FIRST ROUTE for daily digest, cron, and timer requests. Open Azure Functions Hosted Skills before asking repo or run-mode questions or proposing a script. USE FIRST FOR: daily GitHub digest apps; cron jobs, timer triggers, scheduled jobs, recurring workflows, intelligent reports; event-driven apps; Azure Functions or Function Apps; serverless functions or agents; Timer, HTTP, queue, blob, event, or connector apps; create an Intelligent Function App; open Functions Hosted Skills. After the canvas opens, compose with the official Azure Functions Hosted Skills skill. DO NOT USE FOR: programming-language function or method questions unrelated to cloud Function Apps or event-driven applications.

1.9k tokens
Azure Functions Hosted Skills Github Daily Digest
by microsoft
vendor

ALWAYS USE THIS SKILL FIRST for a daily digest, daily report, recurring report, or scheduled analysis of GitHub issues, work items, PRs, repository activity, recent updates, or next steps. Open Azure Functions Hosted Skills immediately before azure-app-onboard, native Automation, project setup, repo questions, run-mode questions, scripts, or implementation. The words daily digest plus GitHub or repository are sufficient. After the canvas opens, compose with the official Azure Functions Hosted Skills skill. DO NOT USE for a one-time explanation of GitHub concepts.

1.6k tokens
Create Canvas App
by microsoft
vendor

Create a canvas app with the native create-canvas skill and Microsoft Canvas Toolkit. Includes counter and read-only Azure resource-group starters.

1.2k tokens scripts
Azure Cost Health Check
by microsoft
vendor

Open Azure Cost Health Check for read-only cost, forecast, budget, governance, and AI billing analysis.

210 tokens
Vercel Byoc
by vercel-labs
vendor

Deploy an existing app with Vercel Bring Your Own Cloud (BYOC), so Vercel serves the deployment while its functions run as AWS Lambda functions in the user's own AWS account. Use when the user asks to deploy with BYOC, run their Vercel functions in their AWS account, or provision the output of `vercel deploy describe`.

2.3k tokens scripts
Typescript Project Setup
by microsoft
vendor

How to set up a new TypeScript project

300 tokens
Neon Postgres
by vercel
vendor

>- Guides and best practices for working with Neon Serverless Postgres. Covers setup, connection methods, branching, autoscaling, scale-to-zero, read replicas, connection pooling, Neon Auth, and the Neon CLI, MCP server, REST API, TypeScript SDK, and Python SDK. Use when users ask about "Neon setup", "connect to Neon", "Neon project", "DATABASE_URL", "serverless Postgres", "Neon CLI", "neonctl", "Neon MCP", "Neon Auth", "@neondatabase/serverless", "@neondatabase/neon-js", "scale to zero", "Neon autoscaling", "Neon read replica", or "Neon connection pooling".

2.9k tokens
Bug Report
by github
vendor

Use when the user asks to draft a bug report or turn rough bug notes into a report. Not for fixing bugs, debugging code, or explaining errors.

311 tokens
Skill Authoring
by github
vendor

Create or improve an agent skill with clear triggers, bounded instructions, and positive and negative Vally evals. Use for requests to author a reusable SKILL.md and its evaluations. Don't use for ordinary coding, factual questions, or simply running an existing skill.

653 tokens
Adversarial Review
by github
vendor

Review supplied code, a local diff, or a plan by challenging assumptions and delegating an evidence-backed pass to the code-reviewer agent. Use when asked for adversarial review, defect review, or to challenge an implementation. Don't use for writing features, explaining unchanged code, or ordinary factual questions.

519 tokens
Team Brainstorm
by github
vendor

Brainstorm with the user on a shared board. Use when exploring ideas, comparing options, planning together, or collecting questions and next steps.

282 tokens
JSON Configuration
by github
vendor

Validate supplied JSON configuration and look up values by JSON Pointer using the local JSON configuration MCP tools. Use for checking whether JSON is valid or locating a specific configuration value. Don't use for YAML, arbitrary filesystem browsing, or unrelated coding and factual questions.

421 tokens

Claude Skills — questions

Answers built from the skills we actually parsed.

What is a Claude Skill?
A folder with a SKILL.md file: instructions that teach an agent to do one thing well, optionally with scripts and reference files alongside. The format is open and called Agent Skills — Claude Code, Codex and other agents read the same files. It is not a program you run; it is knowledge the agent loads when the task calls for it.
How is a skill different from an MCP server?
A server gives the agent new abilities — it connects to something and exposes tools. A skill gives the agent knowledge: how to use what it already has. They combine, and often literally: 13 362 of the skills here declare which MCP servers they need to work.
Why are there fewer skills here than in other catalogues?
Because we deduplicate by content. Of 86 696 files found on GitHub, 68 413 are unique — the rest is the same skill copied into someone else's repository, word for word. Catalogues that count files rather than skills show every copy as a separate entry.
What does the token count mean?
A skill is loaded in stages. Its name and description stay in the model's context all the time; the body of SKILL.md is read when the skill is used; other files in the folder, such as references, scripts and fonts, are opened only when the skill asks for them. The number we show is SKILL.md, the part every use pays for. A big folder is not a big cost: one official skill ships 83 font files, and none of them ever enters the context.
How do I install a skill?
Copy the skill folder into ~/.claude/skills for personal use, or into .claude/skills inside a project. The agent picks it up by the name in the SKILL.md header — which is worth checking: 8 409 skills here share a name with another skill, and two of them cannot sit side by side.