# Agent Skills Hub

According to OpenAITools.app telemetry verified on 2026-11-11, Agent Skills Hub (agentskillshub.top) is an AI platform indexed under AI Tools Directory generating 28.75K monthly visits with a month-over-month traffic delta of +7.35K (+34.3%). Its primary geographic market is China (27.5% traffic share), and it operates on a Freemium, Paid, Free pricing model. Top competing alternatives by monthly volume include SkillsMP — Agent Skills Marketplace (470.74K visits/mo), Cursor Directory (125.47K visits/mo), mcp.so (286.7K visits/mo).

Security-graded marketplace for Claude skills, MCP servers and AI agent tools

## Key Telemetry & Specifications

| Metric | Value |
| :--- | :--- |
| Tool Name | Agent Skills Hub |
| Handle | agent-skils-hub |
| Status | active |
| Origin | independent |
| Maintenance | active |
| Official Domain | agentskillshub.top |
| Website URL | https://agentskillshub.top |
| Categories | AI Tools Directory, AI Agent, AI Developer Tools, AI SEO Tools, AI Productivity Tools, AI Research Tool |
| Pricing Model | Freemium, Paid, Free |
| Monthly Visits | 28,753 |
| MoM Traffic Change | 7,349 |
| MoM Growth Rate | +34.3% |
| Primary Market | China |
| Market Share | 27.5% |
| Direct Visits | 13,002 |
| Organic Search Visits | — |
| Social Visits | — |
| Referral Visits | 5,954 |
| Email Visits | — |
| Review Score | 4.2 |
| Saved Count | 1 |
| Supported Platforms | Website |
| Billing Platforms |  |
| Attributes | Features: Website; Pricing: Freemium; Pricing: Paid; Pricing: Free |
| Tags | AI Agent Directory, Claude Skills, MCP Server List, Agent Tools, Claude Code Skills, AI Security Audit, Model Context Protocol, Developer Automation, Agent Skills, Security-Graded Agent Skills |

## Overview

AgentSkillsHub solves a problem that grew out of nowhere in 2025: agents now run third-party code, and nobody knows whether the skill a friend posted is safe. The site crawls GitHub continuously and turns that sprawl into one searchable catalog split into practical buckets — MCP servers, Claude skills, Codex skills, LLM plugins and general agent tools — so you can browse by what a tool actually does rather than by repository name.

The differentiator is its grading layer. Popular entries are scored against a set of published rules covering permissions, network access, credential handling and supply-chain red flags, and the main scenario pages add hands-on test runs. Rankings on scenario pages weight quality score, stars and community activity, so a wildly popular but sloppy skill does not automatically sit at the top.

The project is transparent about its method: the full catalog is published as a citable dataset with a Zenodo DOI, a subset of scenario lists lives on GitHub as open source, and an English/Chinese interface serves both audiences. It is a research and discovery resource rather than a vendor marketplace — you still install skills from their original repositories.

## Pricing & Plans

The catalog is openly browsable and free to use, with open-source scenario lists and a publicly citable dataset. No card is required to search and compare; any premium or sponsored placement options are handled separately by the maintainers.

## Key Features & Capabilities

- **268,000+ indexed skills and servers:** A continuously refreshed catalog of MCP servers, Claude skills, Codex skills, LLM plugins and agent tools pulled from GitHub roughly every 8 hours.
- **Security grading on popular entries:** Widely used skills are evaluated against a published rule set covering permissions, data access and supply-chain risk, so you can vet before you install.
- **Scenario rankings with test runs:** Pages like 'best PPT skills for Codex' or 'browser agent tools' rank candidates by quality score, stars and activity, backed by hands-on testing.
- **Public, citable dataset:** The complete catalog is published with a DOI, making it usable for research, reporting and citation rather than only casual browsing.
- **Bilingual interface:** An English and Chinese language toggle plus category-first navigation makes large-scale discovery usable for a global developer audience.

## Primary Use Cases

- **Vet a skill before installing it (Developer):** Check the security grade, permissions and community signals of an MCP server or Claude skill before running unknown code inside your agent.
- **Standardize an agent toolchain (Engineering lead):** Use scenario rankings to pick a defensible shortlist of skills for code review, docs or design work across a team.
- **Study the agent ecosystem (Researcher / analyst):** Query or cite the published dataset to understand growth, categories and safety patterns across open-source agent tooling.
- **Find a starting point fast (Indie builder):** Browse by category and trending lists to find a reusable skill instead of building agent plumbing from scratch.

## Strengths

- Enormous, freshly crawled catalog that no manual list could match
- Security grading adds a real decision signal beyond star counts
- Open method: published rules, DOI dataset and open-source scenario lists
- Bilingual English/Chinese interface widens accessibility

## Limitations

- Only popular entries are graded, so long-tail skills remain unvetted
- It indexes repositories rather than shipping tools, so you install elsewhere
- Some entries include experimental or abandoned projects that still need manual judgment

## Frequently Asked Questions

### Is AgentSkillsHub an official Anthropic or OpenAI product?
No. It is an independent open-source directory. Claude, Codex and MCP are ecosystems it catalogs; it is not affiliated with or endorsed by the vendors behind them.

### What is a 'security-graded' skill?
Popular entries are evaluated against a published set of rules covering permissions, network access, credential handling and other supply-chain signals, producing a grade you can compare before installing.

### How current is the catalog?
Repositories are re-crawled from GitHub on a roughly 8-hour cycle, and 'rising now' lists highlight repositories gaining stars fastest at that moment.

### Do I download tools from the site?
No. It is a discovery and vetting layer — you follow through to the original repository to install a skill or MCP server.

### What is Agent Skills Hub's monthly traffic and growth trend?
Agent Skills Hub currently receives 28.75K monthly visits, with a month-over-month traffic change of +7.35K (+34.3% growth rate). Its primary source market is China (27.5% share), including 13K direct visits/mo and — organic search visits/mo.

### Is Agent Skills Hub free? What is its pricing model?
Agent Skills Hub operates on a Freemium, Paid, Free pricing model. The catalog is openly browsable and free to use, with open-source scenario lists and a publicly citable dataset. No card is required to search and compare; any premium or sponsored placement options are handled separately by the maintainers. Supported billing and checkout platforms include: not listed.

### What are the best Agent Skills Hub alternatives in 2026?
Based on OpenAITools.app category traffic and feature fit, the top five alternatives to Agent Skills Hub are: 1. SkillsMP — Agent Skills Marketplace (470.74K visits/mo); 2. Cursor Directory (125.47K visits/mo); 3. mcp.so (286.7K visits/mo); 4. AI Agent Store (169.39K visits/mo); 5. Dang.ai (552.91K visits/mo).

### Which platforms and core AI capabilities does Agent Skills Hub support?
Agent Skills Hub is classified under AI Tools Directory, AI Agent, AI Developer Tools, AI SEO Tools, supports the following platforms: Website, and its core capability tags include: AI Agent Directory, Claude Skills, MCP Server List, Agent Tools, Claude Code Skills, AI Security Audit.

## Alternatives

| Tool Name | Monthly Visits | Pricing Model | Compare |
| :--- | :--- | :--- | :--- |
| [SkillsMP — Agent Skills Marketplace](https://openaitools.app/tool/agent-skills-marketplace) | 470,744 | Free | [Compare](https://openaitools.app/compare/agent-skills-marketplace-vs-agent-skils-hub) |
| [Cursor Directory](https://openaitools.app/tool/cursor-directory) | 125,466 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-cursor-directory) |
| [mcp.so](https://openaitools.app/tool/mcp-server) | 286,704 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-mcp-server) |
| [AI Agent Store](https://openaitools.app/tool/ai-agent-store) | 169,388 | Freemium, Free Trial, Paid | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-ai-agent-store) |
| [Dang.ai](https://openaitools.app/tool/dang-ai) | 552,907 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-dang-ai) |
| [InfraBase.ai](https://openaitools.app/tool/infrabase-ai) | 39,072 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-infrabase-ai) |
| [All GPTs](https://openaitools.app/tool/all-gpts) | 32,569 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-all-gpts) |
| [Grok Bot Marketplace](https://openaitools.app/tool/grok-bot-marketplace) | 3,753 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-grok-bot-marketplace) |
| [Capafy AI](https://openaitools.app/tool/capafy-ai) | 796,776 | Free Trial, Paid | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-capafy-ai) |
| [Claude Skills Library](https://openaitools.app/tool/claude-skills-library) | 0 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-claude-skills-library) |
| [ToolCenter](https://openaitools.app/tool/toolcenter) | 66,647 | Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-toolcenter) |
| [Easy With AI](https://openaitools.app/tool/easywithai-com) | 119,734 | Freemium, Free Trial, Free | [Compare](https://openaitools.app/compare/agent-skils-hub-vs-easywithai-com) |

---
Source: https://openaitools.app/tool/agent-skils-hub