llmref

Versioned documentation registry and semantic search for AI tools and coding assistants.

我该使用它吗

质量与安全性

A
描述质量
95%
模式完整度
98%
命名质量
95%
投毒风险
100%
权限匹配度
100%
协议合规性
100%

基于对工具定义和协议合规性的自动分析。

上下文开销

~652token 数(工具定义)
~1.2 KB典型响应大小
对注意力有中等影响(占 128k 上下文窗口的 0.51%)

这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高,可用于其他任务的注意力就越少。

安装

一键安装

将以下内容添加到你的 `claude_desktop_config.json` 文件中:

{
  "mcpServers": {
    "llmref": {
      "url": "https://api.llmref.org/v1/mcp"
    }
  }
}

远程端点

https://api.llmref.org/v1/mcpstreamable-http

它能做什么

工具清单

工具(4)

🟢 只读🟡 写入🔴 删除⚪ 未知
🟢list_projects(limit, query, skip)

List indexed documentation projects. Use this first to discover what's available. Supports a forgiving fuzzy `query` over slugs (ranked by relevance) plus `limit` and `skip` for pagination.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "default": null,
      "description": "Maximum number of results to return (default 50, max 200).",
      "format": "uint",
      "minimum": 0,
      "type": [
        "integer",
        "null"
      ]
    },
    "query": {
      "default": null,
      "description": "Forgiving fuzzy filter over project slugs; results are ranked by relevance.",
      "type": [
        "string",
        "null"
      ]
    },
    "skip": {
      "default": null,
      "description": "Number of results to skip, for pagination (default 0).",
      "format": "uint",
      "minimum": 0,
      "type": [
        "integer",
        "null"
      ]
    }
  },
  "$schema": "https://json-schema.org/draft/2020-12/schema"
}
🟢list_versions(project)

List available versions for a project. Call this before resolve_docs or search_docs to see valid version strings.

输入模式

{
  "type": "object",
  "properties": {
    "project": {
      "description": "Project slug (e.g. \"nextjs\", \"react\")",
      "type": "string"
    }
  },
  "required": [
    "project"
  ],
  "$schema": "https://json-schema.org/draft/2020-12/schema"
}
🟢resolve_docs(project, version)

Resolve a project version and get artifact URLs (llms.txt, llms-full.txt, meta.json, dumps). Use list_versions first to discover valid version strings.

输入模式

{
  "type": "object",
  "properties": {
    "project": {
      "description": "Project slug (e.g. \"nextjs\", \"react\")",
      "type": "string"
    },
    "version": {
      "description": "Version string or \"latest\"",
      "type": "string"
    }
  },
  "required": [
    "project",
    "version"
  ],
  "$schema": "https://json-schema.org/draft/2020-12/schema"
}
🟢search_docs(project, query, token_budget, top_k, version)

Semantic search across a project's documentation. Returns chunks ranked by relevance. Requires the server to have semantic search enabled.

输入模式

{
  "type": "object",
  "properties": {
    "project": {
      "description": "Project slug (e.g. \"nextjs\", \"react\")",
      "type": "string"
    },
    "query": {
      "description": "Natural-language search query",
      "type": "string"
    },
    "token_budget": {
      "default": 4096,
      "description": "Max tokens to budget for results (default 4096)",
      "format": "uint",
      "minimum": 0,
      "type": "integer"
    },
    "top_k": {
      "default": 10,
      "description": "Max results (1-50, default 10)",
      "format": "uint",
      "minimum": 0,
      "type": "integer"
    },
    "version": {
      "description": "Version string or \"latest\"",
      "type": "string"
    }
  },
  "required": [
    "project",
    "version",
    "query"
  ],
  "$schema": "https://json-schema.org/draft/2020-12/schema"
}

推荐提示词

search_research
Search for information about [topic] using llmref
预期工具: search_docs
find_specific
Find [specific item] using llmref
预期工具: search_docs
list_items
List all [items] available in llmref
预期工具: list_projects
browse_collection
Show me the [collection] from llmref
预期工具: list_projects

社区

评价此服务器

证据

最近观测

已验证未记录版本4 个工具
已验证未记录版本4 个工具