SKILLS & AGENTS/Skill 与 Agent·BROWSER

tandem-browser

hydro13/tandem-browser

AI-Human symbiotic browser — browse the web as one entity

打开仓库Open repository 官网Homepage

怎么装：Install: git clone https://github.com/hydro13/tandem-browser.git

02 / 现在的位置02 / Why now

它现在处在什么位置Where it stands right now

首次发现FIRST SEEN

03 OCT 202603 OCT 2026

增长GROWTH

+0.0%+0.0%

616 → 616 stars · 2 个快照616 → 616 stars · 2 snapshots

状态STATUS

持续活跃Active

03 / 它能帮你做什么03 / What it helps you do

它能帮你做成什么What it helps you do

04 / 安装04 / Install

怎么装How to install

GIT-CLONE

$

README 里给出的安装方式。The installation method given in the README.

MCP CONFIG

$

MCP 客户端配置，粘贴到对应的配置文件里即可。An MCP client config — paste it into the relevant config file.

两种装法选一个即可 —— 上面那条给命令行用户，下面那条给写代码的用户。Pick one of the two — the first is for command-line users, the second for people writing code.

05 / 限制与风险05 / Limits and risk

限制与风险Limits and risk

风险不是警告，是可信度的一部分。以下结论只基于文档静态扫描，我们不会执行项目里的任何代码。Risk here is evidence, not an alarm. These findings come from static scanning of the docs; we never execute a project’s code.

限制LIMITS

需检查Needs review

01需要 API KeyREQUIRES API KEY: 未在文档中发现Not found in the docs
02需要本地环境NEEDS LOCAL RUNTIME: 是Yes
03需要 DockerNEEDS DOCKER: 未在文档中发现Not found in the docs
04难度DIFFICULTY: 需要本地环境Needs a local runtime

PROVENANCE · 这些结论的依据是什么Where these conclusions come from

来源 URLSource URL: https://github.com/hydro13/tandem-browser
原始数据哈希Raw data hash: ed75b315ee1fa6c5eac7846544b7f30a
解析时间Parsed at: 2026-10-03T23:15:23.819Z
最后一次验证Last verified: 2026-10-04T08:53:33.891Z
增长测量时间Growth measured: 2026-10-03T23:24:59.911Z
发现查询Discovery queries: focus:web:code:nextjs
分类来源Category source: rules
焦点Focus: web (3)

静态扫描STATIC SCAN

01控制浏览器Controls a browserreadme
```
## Why Not Just Use Playwright?
```

02控制浏览器Controls a browserskill/SKILL.md

**Caveat — network logs start from DevTools attach time**: the CDP network

03安装时执行脚本Runs a script during installpackage.json
```
postinstall: electron-rebuild -f -w better-sqlite3
```

这里只做静态扫描：读 README、SKILL.md 和依赖清单，不执行代码。没有命中不代表安全。This is static scanning only: we read the README, SKILL.md and dependency list, and never execute code. No findings does not mean safe.

完整解析证据 JSONFull parse evidence JSON

{
  "agents": [
    {
      "agent": "claude-code",
      "evidence": "readme: …xploit. Connect via **MCP** (Claude Code, Claude Desktop, Cursor, Wind…"
    },
    {
      "agent": "claude-desktop",
      "evidence": "readme: …ect via **MCP** (Claude Code, Claude Desktop, Cursor, Windsurf, Ollama, an…"
    },
    {
      "agent": "windsurf",
      "evidence": "readme: …Code, Claude Desktop, Cursor, Windsurf, Ollama, any MCP client) or a…"
    }
  ],
  "apiKey": null,
  "docker": null,
  "taxonomy": {
    "scores": {
      "design": 0,
      "ai-tools": 4,
      "dev-tools": 0,
      "indie-web": 0,
      "skill-agent": 3,
      "productivity": 0
    },
    "primary": "skill-agent",
    "secondary": "browser"
  },
  "localRuntime": "el (Claude, Cursor, a local Ollama model) at a fresh Tandem install and a",
  "skillMdTotal": 1,
  "skillMdErrors": [],
  "treeTruncated": false,
  "categoryScores": {
    "data": 0,
    "agent": 0,
    "media": 0,
    "design": 1,
    "browser": 0,
    "devtool": 0,
    "security": 1,
    "marketing": 0,
    "productivity": 1
  },
  "classification": [
    {
      "signal": "1 SKILL.md with name + description",
      "weight": 0.95
    },
    {
      "signal": "package.json depends on @modelcontextprotocol/sdk + MCP framing",
      "weight": 0.8
    },
    {
      "signal": "README describes a Claude / Agent Skill",
      "weight": 0.7
    },
    {
      "signal": "description mentions AI",
      "weight": 0.2
    }
  ],
  "scannedSources": [
    "readme",
    "skill/SKILL.md"
  ],
  "skillMdFetched": 1
}