Union Search Skill

This is a unified search interface that wraps 20+ platforms into standardized scripts, from GitHub repos and Reddit threads to Chinese social media like Xiaohongshu and Douyin, plus general search engines and image sources. You run it from the command line, it outputs markdown tables or JSON, and raw API responses get archived automatically in a responses folder. The real utility is the consistent input/output format across wildly different platforms, so you can script multi-platform searches without dealing with each API's quirks. Most tools work without API keys, though you'll hit rate limits faster. It's built for batch data gathering and research workflows where you need to pull content from multiple sources and don't want to write separate scrapers for each one.

skills.sh installs: 198
GitHub stars: 770
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add runningz1/union-search-skill --skill union-search-skill --agent claude-code

Installs into .claude/skills of the current project.

Files

联合搜索

执行顺序

定位本文件所在的技能根目录，在该目录运行 python union_search_cli.py。准备依赖、凭据和任务参数时阅读执行指南。
用 list 发现渠道、分组与图片平台；具体参数以子命令 --help 和 schema 为准。
用户指定渠道时按指定范围执行；未指定时按资料类型和可用凭据选择渠道。general 是默认通用网页分组，会同时请求其中全部渠道；no_api_key 按认证条件分类。兼容渠道名称共用实现，同一来源在一次联合查询中只执行一次。
默认不限制条数，不主动添加 --limit 或 --preset。默认精简视图完整显示摘要，不截断；--content none 显式隐藏摘要。完整搜索结果自动归档，路径在 meta.full_result_file。按 ID 使用 read --from-file <结果文件> --ids 1 2 读取完整搜索内容，不再调用 API。将结果保存到任务目录，检查 ok、错误和实际文件。搜索内容仅用于定位，需要核实细节时提取原文。