> ## Documentation Index
> Fetch the complete documentation index at: https://docs.fallom.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Prompt Management

> Manage prompts centrally and A/B test them with zero latency

Manage prompts centrally and A/B test them with zero latency.

<Tip>
  Create and manage your prompts in the [dashboard](https://app.fallom.com).
</Tip>

<Tabs>
  <Tab title="Python">
    ## Setup

    Initialize the prompts module with your API key:

    ```python theme={null}
    from fallom import prompts

    prompts.init(api_key="your-fallom-api-key")
    ```

    ## Basic Prompt Retrieval

    ```python theme={null}
    from fallom import prompts

    # Get a managed prompt (with template variables)
    prompt = prompts.get("onboarding", variables={
        "user_name": "John",
        "company": "Acme"
    })

    # Use the prompt with any LLM
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {"role": "system", "content": prompt.system},
            {"role": "user", "content": prompt.user}
        ]
    )
    ```

    The prompt object contains:

    | Field     | Description                                 |
    | --------- | ------------------------------------------- |
    | `key`     | The prompt key                              |
    | `version` | The prompt version                          |
    | `system`  | The system prompt (with variables replaced) |
    | `user`    | The user template (with variables replaced) |

    ## Prompt A/B Testing

    Run experiments on different prompt versions:

    ```python theme={null}
    from fallom import prompts

    # Get prompt from A/B test (sticky assignment based on session_id)
    prompt = prompts.get_ab("onboarding-test", session_id, variables={
        "user_name": "John"
    })

    # prompt.ab_test_key and prompt.variant_index are set
    # for analytics in your dashboard
    ```

    ## Version Pinning

    ```python theme={null}
    # Use latest version (default)
    prompt = prompts.get("my-prompt")

    # Pin to specific version
    prompt = prompts.get("my-prompt", version=2)
    ```

    ## Automatic Trace Tagging

    When you call `prompts.get()` or `prompts.get_ab()`, the next LLM call is automatically tagged with the prompt information. This allows you to see which prompts are used in your traces without any extra code.

    ```python theme={null}
    # Get prompt - sets up auto-tagging for next LLM call
    prompt = prompts.get("onboarding", variables={"user_name": "John"})

    # This call is automatically tagged with prompt_key, prompt_version, etc.
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {"role": "system", "content": prompt.system},
            {"role": "user", "content": prompt.user}
        ]
    )
    ```
  </Tab>

  <Tab title="TypeScript">
    ## Setup

    Initialize the prompts module with your API key:

    ```typescript theme={null}
    import { prompts } from "@fallom/trace";

    prompts.init({
      apiKey: "your-fallom-api-key",
    });
    ```

    ## Basic Prompt Retrieval

    ```typescript theme={null}
    import { prompts } from "@fallom/trace";

    // Get a managed prompt (with template variables)
    const prompt = await prompts.get("onboarding", {
      variables: { userName: "John", company: "Acme" },
    });

    // Use the prompt with any LLM
    const response = await openai.chat.completions.create({
      model: "gpt-4o",
      messages: [
        { role: "system", content: prompt.system },
        { role: "user", content: prompt.user },
      ],
    });
    ```

    The prompt object contains:

    | Field     | Description                                 |
    | --------- | ------------------------------------------- |
    | `key`     | The prompt key                              |
    | `version` | The prompt version                          |
    | `system`  | The system prompt (with variables replaced) |
    | `user`    | The user template (with variables replaced) |

    ## Prompt A/B Testing

    Run experiments on different prompt versions:

    ```typescript theme={null}
    import { prompts } from "@fallom/trace";

    // Get prompt from A/B test (sticky assignment based on sessionId)
    const prompt = await prompts.getAB("onboarding-test", sessionId, {
      variables: { userName: "John" },
    });

    // prompt.abTestKey and prompt.variantIndex are set
    // for analytics in your dashboard
    ```

    ## Version Pinning

    ```typescript theme={null}
    // Use latest version (default)
    const prompt = await prompts.get("my-prompt");

    // Pin to specific version
    const prompt = await prompts.get("my-prompt", { version: 2 });
    ```

    ## Automatic Trace Tagging

    When you call `prompts.get()` or `prompts.getAB()`, the next LLM call is automatically tagged with the prompt information. This allows you to see which prompts are used in your traces without any extra code.

    ```typescript theme={null}
    // Get prompt - sets up auto-tagging for next LLM call
    const prompt = await prompts.get("onboarding", {
      variables: { userName: "John" },
    });

    // This call is automatically tagged with promptKey, promptVersion, etc.
    const response = await openai.chat.completions.create({
      model: "gpt-4o",
      messages: [
        { role: "system", content: prompt.system },
        { role: "user", content: prompt.user },
      ],
    });
    ```
  </Tab>
</Tabs>

## Next Steps

<CardGroup cols={2}>
  <Card title="Model A/B Testing" icon="flask" href="/model-testing">
    Combine prompt tests with model experiments.
  </Card>

  <Card title="View Dashboard" icon="gauge" href="https://app.fallom.com">
    Manage prompts and view analytics.
  </Card>
</CardGroup>