SkillHub is a registry and trust layer for reusable AI agent skills. Browse skills, inspect permissions and manifests, then run approved skills through controlled APIs or MCP.
SkillHub supports public discovery and inspection today; runtime execution is available behind signed-in Project Keys.
Research a web topic and return concise findings with source URLs.
Codex
Claude Code / MCP
Gemini CLI
GitHub Copilot
OpenClaw
Hermes Agentsk_project_************Use the public catalog for discovery, sign in before runtime work, and keep operator-only workflows behind direct links.
Search public Skills and inspect safe contract metadata.
Inspect schema, permissions, runtime target, and version state.
Run REST / MCP calls only after workspace setup and policy checks.
Paid listings use configured checkout, ledger, refund, and payout controls.
From public discovery to runtime calls, every step keeps contract, permission, and audit context visible.
Find the right Skill for your Agent.
Review manifests, permissions, and runtime details.
Create a workspace project and generate a Project Key.
Call the skill via REST API or MCP with your key.
Track usage, logs, and performance.
For launch operations, do not ask buyers to browse hundreds of listings from scratch. Start them from three proven business jobs, then let the marketplace filters handle the long tail.
Audit AI-search visibility, turn findings into briefs, and build a repair queue that marketing, content, and dev teams can execute.
Best for founders, SEO teams, content operators, and agencies that need visible growth work every week.Improve product titles, PDP quality, listing readiness, review mining, and SKU launch checks before paid traffic arrives.
Best for Shopify stores, Amazon sellers, DTC teams, and agencies preparing campaigns or batch SKU launches.Clean messy spreadsheets, standardize imports, explain metrics, and turn weekly reports into repeatable operator workflows.
Best for operators, finance assistants, growth analysts, and teams that still move work through CSVs and spreadsheets.See declared capabilities and risk level before a skill enters your project.
web.search · Lowweb.fetch · OKdata.store · PassedProject keys, policies, logs, and limits keep runtime usage accountable.
Project Key required · LowRate limit: 60 / min · OKAllowed: web.search, web.fetch · PassedSubmission review separates ready skills from needs-change or blocked listings.
Static analysis · LowPermission review · OKSecurity scan · PassedResearch a web topic and return concise findings with source URLs.
Five value investing masters (Buffett, Munger, Duan Yongping, Sanhu Yi, Li Lu) provide multi-perspective investment analysis with Feynman problem decomposition methodology. Distills 25 core mental models from shareholder letters, books, interviews and public speeches.
五位价值投资大师（李录、段永平、巴菲特、芒格、散户乙）的联合投资辅导系统。以费曼问题拆解法为分析基础，提炼25个核心心智模型，支持单人分析、交叉对比、圆桌讨论三种模式。
A real-format publisher flow check skill that verifies submitted skills can render a detail page before operator approval.
SkillHub keeps reusable Skills discoverable, reviewable, and callable through governed runtime paths, so teams can reduce repeated prompt setup, tool wiring, and handoff work across agent environments.
Reuse Skill contracts for repo analysis, code fixes, release checks, and developer handoffs.
Expose approved Skills through MCP while keeping schemas, scopes, and project controls visible.
Turn research, data, SEO, and operations steps into repeatable workflows for CLI-driven agents.
Give engineering teams reusable checks and workflow context before issues, pull requests, and releases.
Route open agent orchestration through reviewed Skill manifests instead of one-off tool setup.
Capture repeatable agent operations as governed templates teams can inspect and run again.
Explore the public registry first, then connect runtime only after project setup is ready.