Quick Start

Use your existing Claude Max or Claude Pro subscription - no API billing required.

# Initialize with Anthropic
stn init --provider anthropic --ship

# Authenticate with your Claude subscription
stn auth anthropic login

This opens your browser to authorize Station. After authorizing, paste the code and select your model:

✅ Successfully authenticated with Anthropic!

   You're using your Claude Max/Pro subscription.
   Station will automatically refresh tokens as needed.

Select a model for your Claude Max/Pro subscription:

* [1] Claude Opus 4.5
      Most capable model - best for complex tasks

  [2] Claude Opus 4
      Previous Opus version

  [3] Claude Sonnet 4.5
      Balanced performance and speed

  [4] Claude Sonnet 4
      Fast and efficient

  [5] Claude Haiku 4.5
      Fastest model - best for simple tasks

# Set your API key
export OPENAI_API_KEY="sk-..."

# Initialize (defaults to gpt-5-mini)
stn init --provider openai --ship

# Set your API key
export GEMINI_API_KEY="..."

# Initialize
stn init --provider gemini --ship

Use the claude mcp add command:

claude mcp add station -e OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4318 --scope user -- stn stdio

Verify it’s added:

claude mcp list

Add to opencode.jsonc:

{
  "mcp": {
    "station": {
      "enabled": true,
      "type": "local",
      "command": ["stn", "stdio"],
      "environment": {
        "OTEL_EXPORTER_OTLP_ENDPOINT": "http://localhost:4318"
      }
    }
  }
}

Edit your config file:

{
  "mcpServers": {
    "station": {
      "command": "stn",
      "args": ["stdio"],
      "env": {
        "OTEL_EXPORTER_OTLP_ENDPOINT": "http://localhost:4318"
      }
    }
  }
}

Add to .cursor/mcp.json in your project (or ~/.cursor/mcp.json for global):

{
  "mcpServers": {
    "station": {
      "command": "stn",
      "args": ["stdio"],
      "env": {
        "OTEL_EXPORTER_OTLP_ENDPOINT": "http://localhost:4318"
      }
    }
  }
}

You are my Station onboarding guide. Walk me through an interactive hands-on tutorial.

RULES:
1. Create a todo list to track progress through each section
2. At each section, STOP and let me engage before continuing
3. Use Station MCP tools to demonstrate - don't just explain, DO IT
4. Keep it fun and celebrate wins!

THE JOURNEY:

## 1. Hello World Agent
- Create a "hello-world" agent that greets users and tells a joke
- Call the agent and show the result
- Explain what happened behind the scenes
[STOP for me to try it]

## 2. Faker Tools & MCP Templates
- Explain Faker tools (AI-generated mock data for safe development)
- Note: Real MCP tools are added via Station UI or template.json
- Explain MCP templates - they keep credentials safe when deploying
- Create a "prometheus-metrics" faker for realistic metrics data
[STOP to see the faker]

## 3. DevOps Investigation Agent
- Create a "metrics-investigator" agent using our prometheus faker
- It should analyze metrics and identify anomalies
- Call it: "Check for performance issues in the last hour"
[STOP to review the investigation]

## 4. Multi-Agent Hierarchy
- Explain agent hierarchies (coordinators delegate to specialists)
- Create an "incident-coordinator" that delegates to:
  - metrics-investigator (existing)
  - logs-investigator (new - create a logs faker too)
- Show me the hierarchy structure in the .prompt file
- Call coordinator: "Investigate why the API is slow"
[STOP to see delegation in action]

## 5. Inspecting Runs
- Use inspect_run to show detailed execution
- Explain: tool calls, delegations, timing
- Mention Jaeger traces at localhost:16686
[STOP to explore run details]

## 6. Workflow with Human-in-the-Loop
- Create a workflow that:
  1. Runs incident-coordinator to investigate
  2. Switch on severity:
     - Low: auto-remediate
     - High: request human_approval
  3. After approval: generate incident report
- Make it complex (use switch/parallel), not just sequential
- Start the workflow
[STOP for me to approve/reject]

## 7. Evaluation & Reporting
- Run evals on our runs with evaluate_benchmark
- Generate a performance report for our incident team
- Explain what the scores mean
[STOP to review the report]

## 8. Grand Finale
- Direct me to http://localhost:8585 (Station UI)
- Quick tour: Agents, MCP servers, Runs, Workflows
- Celebrate - we built a production-ready incident response system!

## 9. Want More? (Optional)
If I want to continue, briefly explain these advanced features:
- **Schedules**: Cron-based agent scheduling (declarative in .prompt files)
- **Sandboxes**: Isolated Python/Node/Bash code execution for agents
- **Notify Webhooks**: Agents can send alerts to Slack, ntfy, Discord, etc.
- **Bundles**: Package and share your agent teams as portable bundles
- **Deploy**: `stn deploy` to Fly.io, Docker, or Kubernetes
- **Coding Backend**: OpenCode integration for AI-assisted development
- **CloudShip**: Connect to CloudShip for centralized management and team OAuth
Just explain what they do - no need to demo.

Start now with the todo list and Section 1. Make it engaging!