Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.langchain.com/llms.txt

Use this file to discover all available pages before exploring further.

The Managed Deep Agents API is a private preview API for creating, updating, and invoking Managed Deep Agents.
Managed Deep Agents is in private preview. Join the waitlist to request access. For usage guidance, see Deploy Managed Deep Agents.
For an end-to-end workflow with request examples, see Deploy Managed Deep Agents. Browse the Managed Deep Agents API reference pages in the sidebar. The generated reference includes:
  • Agents: Create, list, get, update, and delete Managed Deep Agents.
  • Threads: Create, search, count, and inspect durable thread state.
  • Runs: Start and stream runs on Managed Deep Agent threads.
  • MCP servers: Register MCP servers and upload credentials referenced by agent tools.

Authentication

Requests require a LangSmith API key for a workspace with Managed Deep Agents private preview access. Pass the key with each request:
X-Api-Key: <LANGSMITH_API_KEY>

Base route

The private preview API uses /v1/deepagents. Routes are versioned at /v1/, but the surface is in private preview and may change in backwards-incompatible ways before general availability. See API stability for breaking-change communication.

Resource groups

Resource groupPurpose
AgentsCreate and manage Managed Deep Agent resources.
ThreadsCreate and manage durable thread state for Managed Deep Agents.
RunsStart and stream Managed Deep Agent runs on threads.
MCP serversRegister MCP servers and store credentials referenced by agent tools.

Private preview scope

Managed Deep Agents is available on LangSmith Cloud in the US region only during private preview. Self-hosted and Hybrid deployments are not supported, and EU and other regions will follow general availability. The API also does not mirror every LangSmith Deployment endpoint in private preview. Endpoint groups such as integrations, auth, triggers, skills, and sandboxes are not mirrored yet. For operational notes — supported models, thread retention, rate limits, SDK availability, and support channels during preview — see Limits and notes in the deploy guide.