Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问clear审计提醒

readthedocs-search-apireadthedocs 搜索 API

Agent Skill

用于辅助 API 设计、接口文档、请求响应结构和服务集成说明。它适合让 Agent 梳理 endpoint、生成 OpenAPI 草稿、检查字段命名、整理错误码或辅助前后端联调。使用时需要确认真实业务语义、鉴权方式、分页和错误处理规则;涉及生成接口文档时,应避免凭空补字段,最好从现有代码、schema 或接口样例中提取事实。

总安装

349

周安装

14

GitHub Stars

4

下载量

113
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:readthedocs-search-api(readthedocs 搜索 API)
来源仓库:https://github.com/readthedocs/skills
仓库路径:skills/readthedocs-search-api
安装命令:
npx skills add https://github.com/readthedocs/skills --skill readthedocs-search-api
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/readthedocs/skills --skill readthedocs-search-api

简介

用于辅助 API 设计、接口文档和前后端联调说明。

  • 适合梳理 endpoint、生成 OpenAPI 草稿或检查字段命名。
  • 使用时需确认真实业务语义、鉴权方式和错误处理规则。
  • 涉及生成接口文档时,应避免凭空补字段,最好从现有代码提取事实。
  • 安装前建议确认权限范围和维护状态,以及是否会触发文件读写。

SKILL.md

Read the Docs Search API

Query the public Read the Docs Search API to find documentation across millions of pages. This API is available on both Community and Business sites; the base URL changes by host.

Step-by-step instructions

1. Make a search request

Set RTD_HOST for your site:

  • Community: https://app.readthedocs.org
  • Business: https://app.readthedocs.com

Query the API using:

GET ${RTD_HOST}/api/v3/search/?q={query}&page={page}

Required parameters:

  • q: Your search query (e.g., "authentication", "API pagination")
  • page: Result page number (default: 1)

Note on scoping: You will generally get better and more consistent results by scoping your query to a specific project. Use fielded search syntax inside q, for example project:{project-slug} your terms. Examples: project:docs build, project:sphinx configuration.

2. Parse the response

The API returns JSON with this structure:

{
  "count": 1234,
  "next": "https://readthedocs.org/api/v3/search/?q=query&page=2",
  "previous": null,
  "results": [
    {
      "id": "project-name:doc-title",
      "project": "project-name",
      "version": "latest",
      "title": "Documentation Title",
      "path": "/en/latest/path/to/page.html",
      "domain": "project-name.readthedocs.io",
      "highlight": "snippet of matching <mark>text</mark>...",
      "blocks": [
        {
          "id": "section-id",
          "title": "Section Title",
          "path": "/path#section-id"
        }
      ]
    }
  ]
}

3. Handle pagination

If there are more results, use the next URL to fetch the next page. Continue until next is null.

Examples

Search for authentication documentation

curl "${RTD_HOST}/api/v3/search/?q=authentication"

Search within a specific project

curl "${RTD_HOST}/api/v3/search/?q=project:docs%20authentication"

Project-scoped quick queries

# Sphinx
curl "${RTD_HOST}/api/v3/search/?q=project:sphinx%20configuration"
curl "${RTD_HOST}/api/v3/search/?q=project:sphinx%20autodoc"

# Requests
curl "${RTD_HOST}/api/v3/search/?q=project:requests%20proxies"
curl "${RTD_HOST}/api/v3/search/?q=project:requests%20authentication"

# Read the Docs
curl "${RTD_HOST}/api/v3/search/?q=project:readthedocs%20build"
curl "${RTD_HOST}/api/v3/search/?q=project:readthedocs%20redirects"

Search with pagination

curl "${RTD_HOST}/api/v3/search/?q=API&page=2"

Parse results with Python

import os
import requests

response = requests.get(
    f"{os.environ['RTD_HOST']}/api/v3/search/",
    params={"q": "REST API"}
)

data = response.json()
for result in data['results']:
    print(f"{result['project']}: {result['title']}")
    print(f"  Domain: {result['domain']}")
    print(f"  Path: {result['path']}")

Get all paginated results

def search_all(query):
    all_results = []
    page = 1
    while True:
        response = requests.get(
            f"{os.environ['RTD_HOST']}/api/v3/search/",
            params={"q": query, "page": page}
        )
        data = response.json()
        all_results.extend(data['results'])
        if not data['next']:
            break
        page += 1
    return all_results

Common edge cases

No authentication required: The Search API is public and does not require API keys or authentication.

Rate limiting: The API applies reasonable rate limits. If you receive HTTP 429, implement exponential backoff.

Private documentation excluded: Only publicly available documentation is searchable. Private projects are not included.

Search delay: The search index updates with a slight delay. New documentation may not appear immediately.

Empty results: If a query returns no results, try simpler keywords or browse the project directly on Read the Docs.

Project scoping recommended: The global index is large and queries can be broad. For most use cases, include a project:{project-slug} filter in q to scope results to the relevant documentation project.

Docs

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

github-copilot

24.69%
按下载量换算28

Codex

24.95%
按下载量换算28

continue

16.41%
按下载量换算19

Claude Code

11.55%
按下载量换算13

mcpjam

7.74%
按下载量换算9

roo

3.25%
按下载量换算4

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills