langchain-ai / langchain-ai/langsmith-java
First-class Google ADK integration (parity with Python configure_google_adk)
Nobody has claimed this yet.
- Dominant language
- Kotlin
- Stars
- 33
- Forks
- 8
- Avg merge
- 8d 3h
- Merged PRs (30d)
- 3
Description
We are evaluating Google ADK Java as an in-process alternative to a Python agent service and want to send traces to LangSmith for evaluation.
The Python LangSmith SDK ships a configure_google_adk() integration (langsmith/integrations/google_adk) that automatically instruments Google ADK runners, agents, LLM flows, and tools via wrapt monkey-patching. It produces a fully structured trace hierarchy with:
OpenAI-format messages inputs and tool_calls outputs
llm_call_type classification (tool_selection, response_generation, direct_response)
Token usage mapped to LangSmith's cost schema
Time-to-first-token tracking
Correct dotted_order ancestry across the full agent tree
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the Python reference at langsmith/integrations/google_adk and locate the existing Java SDK integration entry points. Compare the requested Google ADK runner, agent, LLM, and tool instrumentation with the listed trace requirements. Done means parity for structured messages and tool calls, llm_call_type, token usage, time-to-first-token, and dotted_order ancestry.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- kotlin
- Domain
- observability
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 38/100