agentpulse-setup

Set up AgentPulse in a React or Electron project. Use when the user wants to add AgentPulse, make their app MCP-controllable, or integrate with Claude Code/Claude Desktop. This is a ONE-TIME setup skill. For help writing useExpose hooks, use the "agentpulse" skill instead.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "agentpulse-setup" with this command: npx skills add dang-hai/agentpulse/dang-hai-agentpulse-agentpulse-setup

AgentPulse Setup

Set up AgentPulse to make your React app controllable by AI agents via MCP.

Prerequisites

  • Node.js project with React
  • Know if it's a web app (Next.js, Vite, CRA) or Electron app

Step 1: Install

npm install agentpulse

Step 2: Add Provider

Find the app's root component and wrap it with AgentPulseProvider.

Vite / CRA (App.tsx or main.tsx):

import { AgentPulseProvider } from 'agentpulse';

function App() {
  return (
    <AgentPulseProvider endpoint="ws://localhost:3100/ws">
      <YourApp />
    </AgentPulseProvider>
  );
}

Next.js App Router (app/layout.tsx):

'use client';
import { AgentPulseProvider } from 'agentpulse';

export default function RootLayout({ children }) {
  return (
    <html><body>
      <AgentPulseProvider endpoint="ws://localhost:3100/ws">
        {children}
      </AgentPulseProvider>
    </body></html>
  );
}

Next.js Pages Router (pages/_app.tsx):

import { AgentPulseProvider } from 'agentpulse';

export default function App({ Component, pageProps }) {
  return (
    <AgentPulseProvider endpoint="ws://localhost:3100/ws">
      <Component {...pageProps} />
    </AgentPulseProvider>
  );
}

Electron apps: See references/ELECTRON_SETUP.md.

Step 3: Add a Sample useExpose

Add one useExpose call to verify setup works. Find an interactive component:

import { useExpose } from 'agentpulse';

function SomeInput() {
  const [value, setValue] = useState('');

  useExpose('test-input', { value, setValue }, {
    description: 'Test input for verifying AgentPulse setup.',
  });

  return <input value={value} onChange={e => setValue(e.target.value)} />;
}

Step 4: Add npm Scripts

Add to package.json:

{
  "scripts": {
    "agentpulse": "agentpulse"
  }
}

Optional: run app and server together (requires concurrently):

{
  "scripts": {
    "dev:agent": "concurrently \"npm run dev\" \"npm run agentpulse\""
  }
}

Step 5: Configure MCP Client

Claude Desktop

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "agentpulse": {
      "url": "http://localhost:3100/mcp"
    }
  }
}

Restart Claude Desktop after editing.

Claude Code

Add to MCP settings (same JSON format as above).

Step 6: Verify Setup

  1. Start your app: npm run dev
  2. Start AgentPulse: npx agentpulse
  3. In MCP client, run: discover()

You should see your test-input component listed.

Troubleshooting

IssueSolution
"WebSocket connection failed"Is npx agentpulse running?
Components not in discover()Is AgentPulseProvider at root? Refresh browser.
MCP tools not availableCheck config file path, restart MCP client

See references/TROUBLESHOOTING.md for more.

Next Steps

Setup is complete. To expose more components, use the agentpulse skill:

"Help me expose my login form to AgentPulse"

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

agentpulse

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

visual-overlay

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

openclaw-version-monitor

监控 OpenClaw GitHub 版本更新,获取最新版本发布说明,翻译成中文, 并推送到 Telegram 和 Feishu。用于:(1) 定时检查版本更新 (2) 推送版本更新通知 (3) 生成中文版发布说明

Archived SourceRecently Updated
Coding

ask-claude

Delegate a task to Claude Code CLI and immediately report the result back in chat. Supports persistent sessions with full context memory. Safe execution: no data exfiltration, no external calls, file operations confined to workspace. Use when the user asks to run Claude, delegate a coding task, continue a previous Claude session, or any task benefiting from Claude Code's tools (file editing, code analysis, bash, etc.).

Archived SourceRecently Updated