traceloop / traceloop/openllmetry

๐Ÿš€ Feature: Emit OTel Log Events following GenAI semconv for LangChain

Open
#2,322 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement new instrumentation
Dominant language
Python
Stars
7.4k
Forks
1.1k
Avg merge
8d 14h
Merged PRs (30d)
2

Description

Which component is this feature for?

Langchain Instrumentation

๐Ÿ”– Feature description

Hello ๐Ÿ‘‹ The latest OTel Gen AI semantic conventions put prompt content into OTel Log Events instead of Span attributes/events. In particular, it would be great to see this for LangChain/LangGraph instrumentations.

๐ŸŽค Why is this feature needed ?

So telemetry matches the current OTel semantic conventions

โœŒ๏ธ How do you aim to achieve this?

I haven't been joining the GenAI Semconv WG meetings and don't know the current plan. If the plan is to move instrumentation into OTel, I can work on that and implement it in OTel repos.

Otherwise, I can send PRs here.

๐Ÿ”„๏ธ Additional Information

No response

๐Ÿ‘€ Have you spent some time to check if this feature request has been raised before?

There is https://github.com/traceloop/openllmetry/issues/1455 but it doesn't mention log events specifically

  • I checked and didn't find similar issue
Are you willing to submit PR?

Yes I am willing to submit a PR!

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up โ€” it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the linked OTel GenAI semantic conventions for GenAI events and reviewing the existing LangChain and LangGraph instrumentation. Confirm whether the change belongs in this repository or an OTel repository, then define the affected telemetry paths. Done means prompt content is emitted as OTel Log Events in accordance with the current conventions.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai, observability
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.