logo
AgentLed

Quick Start

Install the CLI, connect your MCP client, and run your first workflow in under 5 minutes.

1. Install the CLI

The AgentLed CLI sets up your workspace, configures MCP, and gives you a command-line interface for creating workflows.

# One-command setup (opens browser, creates account, configures MCP)
npx @agentled/cli --setup

# Or install globally
npm install -g @agentled/cli

The setup wizard creates your workspace, generates an API key, and connects your MCP client automatically.

2. Connect MCP

AgentLed works as an MCP server. Connect it to Claude Code, Cursor, Windsurf, Codex, or any MCP-compatible client.

Claude Code

claude mcp add agentled \
  -e AGENTLED_API_KEY=wsk_... \
  -- npx -y @agentled/mcp-server

Cursor / Windsurf

Add this to your MCP settings JSON:

{
  "mcpServers": {
    "agentled": {
      "command": "npx",
      "args": ["-y", "@agentled/mcp-server"],
      "env": {
        "AGENTLED_API_KEY": "wsk_..."
      }
    }
  }
}

3. Get Your API Key

1.Sign up at agentled.app

2.Open Workspace Settings > Developer

3.Generate a new API key (starts with wsk_)

4. Create Your First Workflow

Describe what you want in natural language. AgentLed builds the workflow, selects integrations, and configures each step.

agentled create "Find fintech CTOs in Europe, enrich via LinkedIn + Hunter,
score by ICP fit, draft personalized outreach"

Or use the MCP chat tool from your AI client:

> Use the agentled chat tool to create a workflow that scrapes
  trending topics in AI, generates 5 LinkedIn posts, and schedules them.

5. Run and Monitor

Start workflows via CLI, MCP, or the web dashboard. View results in real-time.

# Start a workflow
agentled start <workflow-id>

# List recent executions
agentled executions list <workflow-id>

# View execution details
agentled executions get <execution-id>

Results persist in the Knowledge Graph. Each run builds on previous outcomes — scoring accuracy improves automatically over time.

Next Steps