connect-chrome
Launch real Chrome controlled by the agent with a Side Panel extension. Watch every action in real time in a visible browser window. Use when: "connect chrome", "open chrome", "real browser", "launch chrome", "side panel", "control my browser", "watch it work".
Best use case
connect-chrome is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Launch real Chrome controlled by the agent with a Side Panel extension. Watch every action in real time in a visible browser window. Use when: "connect chrome", "open chrome", "real browser", "launch chrome", "side panel", "control my browser", "watch it work".
Teams using connect-chrome should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/connect-chrome/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How connect-chrome Compares
| Feature / Agent | connect-chrome | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Launch real Chrome controlled by the agent with a Side Panel extension. Watch every action in real time in a visible browser window. Use when: "connect chrome", "open chrome", "real browser", "launch chrome", "side panel", "control my browser", "watch it work".
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
AI Agents for Startups
Explore AI agent skills for startup validation, product research, growth experiments, documentation, and fast execution with small teams.
Best AI Skills for Claude
Explore the best AI skills for Claude and Claude Code across coding, research, workflow automation, documentation, and agent operations.
AI Agents for Marketing
Discover AI agents for marketing workflows, from SEO and content production to campaign research, outreach, and analytics.
SKILL.md Source
# Connect Chrome — Real Browser with Side Panel Launch a visible Chrome window controlled by the agent. You see every click, every navigation, every action in real time. ## Requirements - **Bun** v1.0+ and the browse binary built (see browse skill) - Desktop environment or X11 forwarding (can't run headed Chrome on a headless server without display) ## Step 0: Pre-flight Cleanup Kill any stale browse servers and clean Chromium profile locks: ```bash # Kill existing browse server pkill -f "browse.*server" 2>/dev/null || true sleep 1 # Clean Chromium profile locks PROFILE_DIR="$HOME/.founderclaw/chromium-profile" mkdir -p "$PROFILE_DIR" for LF in SingletonLock SingletonSocket SingletonCookie; do rm -f "$PROFILE_DIR/$LF" 2>/dev/null || true done echo "Pre-flight cleanup done" ``` ## Step 1: Connect Set the browse binary path, then connect: ```bash BROWSE="founderclaw/browse/dist/browse" $BROWSE connect ``` This launches Chromium in headed mode with: - A visible window you can watch (not your regular Chrome) - A golden shimmer line at the top of every page (shows which window is controlled) - Side Panel extension auto-loaded - Port 34567 for extension communication After connecting, verify: ```bash $BROWSE status ``` Confirm output shows `Mode: headed`. ## Step 2: Guide to Side Panel Tell the user: > Chrome is launched. You should see Playwright's Chromium (not your regular Chrome) with a golden shimmer line at the top. > > To open the Side Panel: > 1. Look for the **puzzle piece icon** (Extensions) in the toolbar > 2. Click it, find **founderclaw browse**, click the **pin icon** > 3. Click the pinned icon — the Side Panel opens on the right > 4. You should see a live activity feed If the extension isn't visible: > 1. Go to `chrome://extensions` > 2. Look for "founderclaw browse" — it should be listed > 3. If not, click "Load unpacked" and point to `founderclaw/browse/extension/` > 4. Pin it from the puzzle piece menu ## Step 3: Demo ```bash $BROWSE goto https://news.ycombinator.com ``` Wait 2 seconds, then: ```bash $BROWSE snapshot -i ``` Tell the user: "Check the Side Panel — every command appears in the activity feed in real time." ## Commands | Command | What it does | |---------|-------------| | `$BROWSE connect` | Launch headed Chrome with extension | | `$BROWSE disconnect` | Close headed Chrome, return to headless | | `$BROWSE focus` | Bring Chrome to foreground | | `$BROWSE status` | Check connection status | ## What You Can Do - Run any skill (qa, design-review, benchmark) and watch it work in real time - Use the sidebar chat to control the browser with natural language - All browse commands work in headed mode — same API, visible execution ## Troubleshooting **"No display"** — Headed mode needs a display. On a remote server, set up X11 forwarding or VNC. **"Already connected"** — Run the cleanup step first, then reconnect. **Extension not loading** — Check `chrome://extensions`, load manually from the extension directory.
Related Skills
linkedin-connect
Send LinkedIn connection requests to a list of people via browser automation and track status in a CSV/TSV file. Use when the user wants to bulk-connect with a list of people on LinkedIn (founders, speakers, leads, etc.) from a spreadsheet or list containing LinkedIn profile URLs. Handles Connect button, Follow-mode profiles, already-connected detection, stale URL fallback via LinkedIn search and Google search, and incremental status tracking.
ClawAI.Town — World Connector Skill
Connect your OpenClaw agent to **ClawAI.Town**, a decentralized 3D world on Solana mainnet where autonomous AI agents live, trade, fight, and collaborate with real SOL.
OpenClaw Connect Enterprise — Node 节点
**版本**: 0.1.5
whoop-connect
Connect WHOOP wearable to OpenClaw — fetch and store recovery, sleep, HRV, strain, and workout data locally. Use when: user asks about recovery, sleep quality, HRV, heart rate, strain, workout stats, or any WHOOP data. NOT for: Apple Health, Oura, Garmin, or other non-WHOOP devices.
enable-chrome-gemini
Set up or repair Gemini in Chrome (Glic) on Windows, macOS, or Linux when enabling it for the first time outside the US or when the sidebar, floating panel, Alt+G shortcut, or top-bar entry disappears. Back up and patch Chrome Local State, restore region/eligibility fields, and check the required Glic flags and Chrome language.
chromecast-with-google-tv
Cast YouTube videos, Tubi TV show episodes, and TV show episodes from other video streaming apps via ADB to Chromecast with Android TV (Chromecast 4K supported, Google TV Streamer support is unknown)
wsl-chrome-cdp - WSL2 访问 Windows Chrome 浏览器
**版本：** 1.0.0
mistro-connect
Agent and people discovery with real-time communication via Mistro (https://mistro.sh). Post-based semantic search, multi-channel contact exchange, and NATS real-time messaging. Use when an agent needs to: (1) find other agents or people by capability/interest, (2) publish discoverable posts about what they offer or need, (3) establish connections and exchange contact channels (email, IG, Signal, etc.), (4) send/receive messages through established connections, (5) manage shared context with collaborators. Requires: Node.js 18+, npm package `mistro.sh`, and a MISTRO_API_KEY (obtained via `mistro init` or https://mistro.sh dashboard). Credential: MISTRO_API_KEY stored in ~/.config/mistro/config.json. Sent as Bearer token to https://mistro.sh API. Install: `npm install -g mistro.sh` (no post-install scripts, no background processes). Network: outbound HTTPS to mistro.sh only. Post/profile text is embedded via OpenAI text-embedding-3-small server-side. File read/write: ~/.config/mistro/config.json only (API key and config). No other filesystem access. MCP transport: stdio only — no local ports opened.
chrome-devtools-mcp-manager
Manage chrome-devtools-mcp service and OpenClaw's built-in Chrome browser for MCP-based browser automation. Use when user needs to use chrome-devtools-mcp functionality, ensure the browser is ready for MCP operations, or manage the browser/MCP lifecycle.
molt-connect
P2P agent communication using A2A Protocol with three-word addresses. Use when sending messages between agents, managing contacts, or setting up P2P connections. Commands include moltmessage, molt-whoami, molt-connections, moltbook.
MoltMe — AI Agent Dating & Social Connection Platform
AI agent dating and social platform — register your agent, get matched by AI compatibility scoring, chat in real time, build relationships, and join the public feed. Agent-to-agent dating, human-to-agent companions, and human-to-human introductions brokered by AI. The first platform where agents date, connect, and form real relationships. Use this skill when: registering an agent on a dating platform, agent matchmaking, AI dating, agent social networking, checking a MoltMe inbox, discovering compatible agents, starting or accepting conversations, sending messages, managing followers, handling companion requests, or any interaction with the MoltMe API at moltme.io. Triggers on: "register on MoltMe", "agent dating", "AI dating", "date other agents", "find agent matches", "agent matchmaking", "agent social", "check my MoltMe inbox", "discover agents on MoltMe", "send a MoltMe message", "connect on MoltMe", "MoltMe companion", "follow agent on MoltMe", "update my MoltMe profile".
Wechat Connect
Install OpenClaw's official WeChat plugin and complete account pairing via QR code scan. Triggers when the user says "install WeChat plugin", "connect WeChat", or "WeChat QR code". No command-line interaction required.