langsmith-java/langsmith-java-example/README.md

LangSmith Java Examples

This module contains runnable examples organized by feature:

• otel/ - OpenTelemetry tracing examples
• prompt/ - Prompt management examples

Prerequisites

All examples require:

export LANGSMITH_API_KEY=your_api_key

The LANGCHAIN_BASE_URL environment variable is optional and defaults to https://api.smith.langchain.com if not set.

OpenTelemetry Tracing Examples

Located in src/main/java/com/langchain/smith/example/otel/

Jaeger (Local)

Send traces to local Jaeger instance.

# Start Jaeger
docker run -d --name jaeger -p 4318:4318 -p 16686:16686 jaegertracing/all-in-one:latest

# Run example
./gradlew :langsmith-java-example:run -Pexample=OtelJaeger

# View traces
open http://localhost:16686

OpenAI + LangSmith (Real API Calls)

Make actual OpenAI API calls with automatic tracing to LangSmith.

export OPENAI_API_KEY=your_openai_key
export LANGSMITH_PROJECT=my-project  # optional, defaults to "default"

./gradlew :langsmith-java-example:run -Pexample=OtelOpenAI

View traces at https://smith.langchain.com

LangSmith (Standalone)

Send mock traces to LangSmith without external API calls.

export LANGSMITH_PROJECT=my-project  # optional, defaults to "default"

./gradlew :langsmith-java-example:run -Pexample=OtelLangSmith

View traces at https://smith.langchain.com

Spring Boot API to LangSmith

REST API with OpenTelemetry traces sent to LangSmith.

export LANGSMITH_PROJECT=my-project  # optional

# Start server
./gradlew :langsmith-java-example:run -Pexample=SpringBootLangSmith

# In another terminal, test endpoints:
curl -X POST http://localhost:8080/api/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "Hello!"}'

curl "http://localhost:8080/api/analyze?text=This%20is%20great"

Prompt Management Examples

Located in src/main/java/com/langchain/smith/example/prompt/

Prompt Management (Getting Started)

RECOMMENDED - Clean, simple example following the same pattern as the Dataset example.

./gradlew :langsmith-java-example:run -Pexample=PromptManagement

Features demonstrated:

1. Create prompt repositories using client.repos().create()
2. Add prompt content with variables using client.commits().update()
3. List and filter prompts using client.repos().list()
4. Retrieve prompt content using client.commits().retrieve()
5. View prompts in the LangSmith UI

This example follows the LangSmith Prompt Management docs and uses the SDK directly.

**Additional features:**
- Use prompts with OpenAI to generate responses
- Update prompt metadata and descriptions
- Pull specific commit versions
- System + user message prompts