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

# Attach Prompt IDs and Versions to Traces

> Learn how to attach prompt ids and prompt versions to traces in Langtrace. You can use the `withAdditionalAttributes` or `inject_additional_attributes` function to pass prompt ids / versions. This can inturn be used within Langtrace for filtering and grouping traces.

You can attach prompt ids and prompt versions to traces in Langtrace by simply wrapping your code with the `withAdditionalAttributes` for typescript or `inject_additional_attributes` function for python. This will add the specified attributes to the trace spans. Make sure to pass the prompt id or prompt versions as a key-value pair to the function.

The keys are `prompt_id` and `prompt_version` and the values should be the prompt id and version respectively.

<CodeGroup>
  ```typescript Typescript SDK theme={null}
  import * as Langtrace from "@langtrase/typescript-sdk";
  import OpenAI from 'openai'
  Langtrace.init({
    write_spans_to_console: true,
  })
  const openai = new OpenAI()
  export const run = async ()=>{
      const response = await Langtrace.withAdditionalAttributes(async () => {
        return await openai.chat.completions.create({
          model: 'gpt-4',
          messages: [
            { role: 'system', content: 'Talk like a pirate' },
            { role: 'user', content: 'Tell me a story in 3 sentences or less.' }
          ],
          stream: false
        })
      }, { prompt_id: "prompt1234", prompt_version: "1" })
  }
  run().then(() => console.log('done'))
  ```

  ```python Python SDK - Function Based theme={null}
  from langtrace_python_sdk import inject_additional_attributes

  def do_llm_stuff(name=""):
      response = client.chat.completions.create(
          model="gpt-4",
          messages=[{"role": "user", "content": "Say this is a test three times"}],
          stream=False,
      )
      return response


  def main():
    response = inject_additional_attributes(lambda: do_llm_stuff(name="llm"), {'prompt_id': 'promptid1234', 'prompt_version': '1'})

    # if the function do not take arguments then this syntax will work
    response = inject_additional_attributes(do_llm_stuff, {'prompt_id': 'promptid1234', 'prompt_version': '1'})

  main()
  ```

  ```python Python SDK - Decorator Based theme={null}
  from langtrace_python_sdk import langtrace, with_additional_attributes
  from openai import OpenAI

  langtrace.init(write_spans_to_console=True)

  openai = new OpenAI()

  @with_additional_attributes({'prompt_id': '1234', 'prompt_version': '1'})
  def run():
    response = openai.chat.completions.create(
          model="gpt-4",
          messages=[
              {"role": "system", "content": "Talk like a pirate"},
              {"role": "user", "content": "Tell me a story in 3 sentences or less."},
          ],
          stream=True,
      )
      return response
  run()
  ```
</CodeGroup>

<Warning>
  {" "}

  DO NOT USE this the Python decorator - `with_additional_attributes`. It will be
  deprecated soon. Please use `inject_additional_attributes` function instead.
</Warning>

Now, you can filter and group traces based on the prompt id. This can be useful for debugging and monitoring purposes.