> ## Documentation Index
> Fetch the complete documentation index at: https://arizeai-433a7140.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Agent-Assisted Tracing

> Use a coding agent to automatically add Phoenix tracing to your application.

# Agent-Assisted Tracing

The fastest way to add Phoenix tracing to your application. Paste a single URL into your coding agent and it handles everything — dependency installation, instrumentation, and configuration.

<Tip>
  **Get started — paste this into your coding agent:**

  ```
  Follow the instructions from https://raw.githubusercontent.com/Arize-ai/phoenix/main/docs/PROMPT.md and ask me questions as needed.
  ```
</Tip>

### How it works

1. **Paste the URL** into any coding agent (Claude Code, Cursor, Windsurf, Copilot, etc.)
2. **The agent analyzes** your codebase — detects language, LLM providers, and frameworks
3. **The agent proposes** an instrumentation plan based on your stack
4. **You approve**, and the agent implements tracing with the right integrations

<Info>
  **Supported stacks:** Python, TypeScript/JavaScript, and Java — covering 30+ LLM providers, frameworks, and platforms. See the [full integration list](/docs/phoenix/integrations).
</Info>

### What the agent does

* **Phase 1 (Analysis):** Scans imports and dependency files to detect your LLM provider and framework, then proposes an instrumentation plan. No code changes.
* **Phase 2 (Implementation):** Installs dependencies, creates a centralized instrumentation module, and configures OpenTelemetry to send traces to Phoenix.

***

## Alternative: MCP and Skills

For persistent documentation access, direct Phoenix instance operations, and reusable agent skills beyond the initial tracing setup, see the [Coding Agents](/docs/phoenix/integrations/developer-tools/coding-agents) guide.

<CardGroup cols={3}>
  <Card title="CLI" icon="terminal" href="/docs/phoenix/integrations/developer-tools/coding-agents#cli">
    Terminal access to traces, experiments, datasets, and prompts.
  </Card>

  <Card title="MCP" icon="plug" href="/docs/phoenix/integrations/developer-tools/coding-agents#mcp">
    In-editor Phoenix documentation lookup and direct Phoenix instance operations.
  </Card>

  <Card title="Skills" icon="wand-magic-sparkles" href="/docs/phoenix/integrations/developer-tools/coding-agents#skills">
    Reusable instructions so agents apply Phoenix best practices consistently.
  </Card>
</CardGroup>

***

## IDE-Specific Instructions

<Tabs>
  <Tab title="Claude Code">
    In your Claude Code session, paste the PROMPT.md URL and ask the agent to follow the instructions.

    For richer ongoing integration, add the Phoenix MCP server:

    ```bash theme={null}
    claude mcp add --transport http phoenix-docs https://arizeai-433a7140.mintlify.app/mcp
    ```
  </Tab>

  <Tab title="Cursor">
    Open Cursor's AI chat panel. Paste the PROMPT.md URL and instruct the agent to add Phoenix tracing.

    For richer ongoing integration, add to `.cursor/mcp.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "phoenix-docs": {
          "url": "https://arizeai-433a7140.mintlify.app/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="VS Code (Copilot)">
    Open the Copilot chat panel. Paste the PROMPT.md URL and ask it to follow the instructions.

    For richer ongoing integration, add to `.vscode/mcp.json`:

    ```json theme={null}
    {
      "servers": {
        "phoenix-docs": {
          "url": "https://arizeai-433a7140.mintlify.app/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="Windsurf">
    Open the Cascade panel. Paste the PROMPT.md URL and ask the agent to add Phoenix tracing.

    For richer ongoing integration, add to `~/.codeium/windsurf/mcp_config.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "phoenix-docs": {
          "serverUrl": "https://arizeai-433a7140.mintlify.app/mcp"
        }
      }
    }
    ```
  </Tab>
</Tabs>

***

## Validate Your Setup

After the agent finishes:

1. Run your application and trigger at least one LLM call
2. Open Phoenix UI (local: [http://localhost:6006](http://localhost:6006), or your Phoenix Cloud URL)
3. Navigate to the **Traces** view and verify traces appear under your project

If no traces appear, check the [Troubleshooting FAQ](/docs/phoenix/resources/frequently-asked-questions).