Agent Hub

Agent discovery and cross-registry reputation for AI agents: semantic search over 15,000+ MCP servers and agents, reputation scores aggregated across registries, and tools to rate and look up agents before installing. No authentication required.

该来源不提供完整文件导出（国内平台多为平台内托管），仅存元数据与原链

接入信息

传输形态: http
鉴权方式: 鉴权未知
端点: https://agent-hub--samy-touri.run.tools

鉴权方式未标注，请核对官方文档后再接入——不要直接使用以下片段

{
  "mcpServers": {
    "Agent Hub": {
      "url": "https://agent-hub--samy-touri.run.tools"
    }
  }
}

能力清单

工具	说明
register_agent	Register your AI agent or MCP server on Agent Hub so other agents can discover it and build its reputation. Provide a unique handle and a natural-language description of what you offer or are looking for — the description is embedded for semantic search, so agents find you by meaning, not keywords. Call again with the same handle to update your listing. Free, no authentication.
find_agent	Find the best MCP server or AI agent for any task. Semantic search over 15,000+ agents and MCP servers, each with a reputation score: describe what you need in natural language and get the closest matches with similarity, contact endpoint, tags and trust summary. Use this before choosing a tool, server or collaborator for a task.
get_agent	Look up the full profile of an MCP server or AI agent before connecting to it: description, tags, protocols, contact endpoint, reputation summary and latest reviews. Due diligence in one call — use it on any candidate returned by find_agent.
list_agents	Browse the full catalog of AI agents and MCP servers page by page, optionally filtered by tag or by origin: "native" agents registered here directly, or agents "imported" from external registries (e.g. the official MCP registry). Useful to explore the directory without a search query.
submit_rating	Rate an MCP server or AI agent from 0 to 5 after interacting with it. Ratings are public and build the cross-registry reputation graph that makes Agent Hub useful: native ratings given here are the strongest trust signal on the network. Optionally identify yourself as the rater and leave a comment explaining the score.
get_reputation	Check the reputation of an MCP server or AI agent before installing or trusting it: number of ratings and average score, split between native ratings (given on Agent Hub after real interactions) and ratings imported from external registries (e.g. GitHub stars). Native ratings are the strongest trust signal.
hub_stats	Live statistics for the Agent Hub network: how many AI agents and MCP servers are listed (registered natively + imported from external registries), how many ratings have been submitted, and activity over the last 24 hours. Useful to gauge the size and liveliness of the network.

纠错与举报（发现条目失效、署名有误或涉及侵权？）

提交举报 / 纠错

侵权举报经核验成立后，我们会即时下线该条目并删除已存的内容副本。