{"id":22097,"plugin_id":"plugins_6ab007edb2e48191a279812a72fb0c25","kind":"skill","collection_source":null,"comparison_source":null,"observed_at":"2026-09-30T23:17:08.531Z","digest":"e34c4d56ed63a8fc5468fb02cf0a4586a7b70a2c9d706dcb8b3d3118af8e257e","against":null,"payload":{"name":"agent-reach","description":"Use when the user asks to research, search, investigate, look up, find sources, compare public information, or discover discussions about a topic, including Portuguese requests such as pesquise, procure fontes, investigue, or pesquisar, across the web, GitHub, Reddit, YouTube, social platforms, articles, RSS, or other public sources.","included_files":[{"relative_path":"SKILL_en.md","size_in_bytes":4960},{"relative_path":"SOURCE-METADATA.yaml","size_in_bytes":98},{"relative_path":"agents/openai.yaml","size_in_bytes":273},{"relative_path":"references/career.md","size_in_bytes":712},{"relative_path":"references/dev.md","size_in_bytes":1383},{"relative_path":"references/search.md","size_in_bytes":821},{"relative_path":"references/social.md","size_in_bytes":7729},{"relative_path":"references/video.md","size_in_bytes":4050},{"relative_path":"references/web.md","size_in_bytes":1103}],"skill_md_contents":"---\nname: agent-reach\ndescription: Use when the user asks to research, search, investigate, look up, find sources, compare public information, or\n  discover discussions about a topic, including Portuguese requests such as pesquise, procure fontes, investigue, or pesquisar,\n  across the web, GitHub, Reddit, YouTube, social platforms, articles, RSS, or other public sources.\ncompatibility: Requires the Agent Reach CLI plus network access and any platform-specific CLIs, authentication, or APIs described\n  in the bundled references.\n---\n\n# Agent Reach — 互联网能力路由器\n\n## Host capability adaptation\n\n1. **Native host capability:** when ChatGPT/Codex exposes a native tool or connected source that satisfies this task, use it.\n2. **Original external runtime:** when the upstream runtime named by this skill is actually available, use it as documented below.\n3. **Instruction-only fallback:** when neither is available, perform only the reasoning/instruction portion that remains valid, state the limitation, and never fabricate tool output, successful execution, or persisted state.\n\nUse native host web, GitHub, or connected-source tools when available; otherwise provide only the instruction-based research plan and clearly disclose the missing runtime.\n\n13 平台、多后端。**本 skill 存在时必须用它访问这些平台，不要自己发明方案。**\n\n## 常驻规则（全程适用）\n\n1. **动手前先体检**：多后端平台（小红书/Reddit/B站/Twitter）先跑\n   `agent-reach doctor --json`，按各平台 `active_backend` 字段选命令组。\n2. **声明你在用什么**：开始干活前说一句「使用 agent-reach 的 X 平台 / Y 后端」。\n3. **失败按 references 里的重试链处理**，不要瞎猜命令。\n4. **全网调研类任务**：组合多平台（Exa 搜索 + Twitter/Reddit 看讨论 + 小红书/B站看中文场景），并行收集再汇总。\n5. **替用户盯版本**：完成一次较大的调研/多平台任务后，顺手跑\n   `agent-reach check-update`（很快，一个 API 调用）。有新版就在收尾汇报里附一句：\n   「Agent Reach 有新版 vX.Y.Z，复制这句话给我即可更新：帮我更新 Agent Reach：\n   https://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/update.md」。\n   不要中断当前任务去更新，也不要重复提醒同一个版本。\n\n## 路由表\n\n| 用户意图 | 分类 | 详细文档 |\n|---------|------|---------|\n| 网页搜索/代码搜索 | search | [references/search.md](references/search.md) |\n| 小红书/推特/B站/V2EX/Reddit | social | [references/social.md](references/social.md) |\n| 招聘/职位/LinkedIn | career | [references/career.md](references/career.md) |\n| GitHub/代码 | dev | [references/dev.md](references/dev.md) |\n| 网页/文章/RSS | web | [references/web.md](references/web.md) |\n| YouTube/B站/播客字幕 | video | [references/video.md](references/video.md) |\n\n## 零配置快速命令\n\n```bash\n# Exa 网页搜索\nmcporter call 'exa.web_search_exa(query: \"query\", numResults: 5)'\n\n# 通用网页阅读\ncurl -s \"https://r.jina.ai/URL\"\n\n# GitHub 搜索\ngh search repos \"query\" --sort stars --limit 10\n\n# YouTube 字幕（注意：B站不要用 yt-dlp，见 video.md）\nyt-dlp --write-sub --skip-download -o \"/tmp/%(id)s\" \"URL\"\n\n# V2EX 热门\ncurl -s \"https://www.v2ex.com/api/topics/hot.json\" -H \"User-Agent: agent-reach/1.0\"\n\n# B站搜索（bili-cli，无需登录）\nbili search \"query\" --type video -n 5\n```\n\n## 需登录态的平台（按 doctor 的 active_backend 选命令）\n\n```bash\n# Twitter 搜索（twitter-cli 首选；失败重试链见 social.md）\ntwitter search \"query\" -n 10\n\n# Reddit（无零配置路径：OpenCLI 或 rdt-cli，必须登录态）\nopencli reddit search \"query\" -f yaml   # 桌面\nrdt search \"query\" --limit 10            # 存量/服务器\n\n# 小红书（桌面首选 OpenCLI）\nopencli xiaohongshu search \"query\" -f yaml\n```\n\n## 环境检查\n\n```bash\n# 检查可用 channel 与每个平台当前激活的后端\nagent-reach doctor --json\n```\n\n## 工作区规则\n\n**不要在 agent workspace 创建文件。** 使用 `/tmp/` 存放临时输出，`~/.agent-reach/` 存放持久数据。\n\n## 详细文档\n\n根据用户需求，阅读对应的详细文档：\n\n- [搜索工具](references/search.md) — Exa AI 搜索\n- [社交媒体](references/social.md) — 小红书, Twitter, B站, V2EX, Reddit（多后端命令组）\n- [职场招聘](references/career.md) — LinkedIn\n- [开发工具](references/dev.md) — GitHub CLI\n- [网页阅读](references/web.md) — Jina Reader, RSS\n- [视频播客](references/video.md) — YouTube, B站, 小宇宙\n\n## 配置渠道\n\n如果某个 channel 需要配置，获取安装指南：\nhttps://raw.githubusercontent.com/Panniantong/agent-reach/main/docs/install.md\n\n用户只需提供 cookies，其他配置由 agent 完成。\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}