Google ADK agents can now deploy to LangSmith with one command
LangChain's LangSmith Deployments now supports Google ADK (Agent Development Kit) agents, adding session persistence, streaming, and tracing via a small wrapper package and a single deploy command.
Score breakdown
Teams building agents with Google ADK gain a path to production-grade managed infrastructure — with persistence, streaming, and tracing — without rebuilding their agent outside the ADK framework.
- 01LangSmith Deployments now supports Google ADK (Agent Development Kit) agents.
- 02The `deployments-wrap-sdk` PyPI package wraps a Google ADK runner into a LangSmith-compatible agent.
- 03A `LangsmithSessionService` provides checkpointing for thread-level memory and short-term persistence.
LangChain published a walkthrough showing how Google ADK (Agent Development Kit) agents can be deployed directly to LangSmith Deployments. The integration is built around `deployments-wrap-sdk`, a wrapper package available on PyPI that converts a standard Google ADK runner into a LangSmith deployment-compatible agent. Developers add a `LangsmithSessionService` to their project to plug into LangSmith's checkpointing system, providing short-term persistence and thread-level memory across conversation turns. From there, the agent deploys with a single command, and session persistence, streaming, and tracing are included automatically.
The demo agent is a personal task manager built with `googleadk.agents`.
The demo agent is a personal task manager built with `googleadk.agents`. It exposes tools for adding tasks (with title and priority), listing tasks, marking tasks as completed, deleting tasks, and summarizing the task list. The agent is defined with a name, model, description, system prompt, and a list of tools — standard Google ADK patterns — before being wrapped with the LangSmith deployment SDK. The walkthrough covers local testing before deployment to LangSmith. The source transcript is truncated before the deployment steps are shown.
Key facts
- 01LangSmith Deployments now supports Google ADK (Agent Development Kit) agents.
- 02The `deployments-wrap-sdk` PyPI package wraps a Google ADK runner into a LangSmith-compatible agent.
- 03A `LangsmithSessionService` provides checkpointing for thread-level memory and short-term persistence.
- 04Deployment includes session persistence, streaming, and tracing out of the box.
- 05The demo is a task-manager agent with tools for adding, listing, completing, deleting, and summarizing tasks.
- 06The agent uses ADK session state to remember the task list across conversation turns.
- 07Deployment is triggered via a single deploy command after local testing.
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 8, 2026 · 15:36 UTC. How this works →