managed-deepagents 0.5.1.dev101

[!IMPORTANT] Public beta. Managed Deep Agents is in public beta. The PyPI package API and managed runtime contract may change. See the docs for getting started and updates.

managed-deepagents is the PyPI package for authoring Managed Deep Agents in Python. It includes:

• define_deep_agent, the Python authoring contract for managed agents.
• define_schedule, the Python contract for managed cron schedules.
• mda, the CLI used to build and deploy your agent to LangSmith.
• managed_deepagents.runtime, the runtime helper used by generated managed entry modules.

Install

uv tool install managed-deepagents

[!NOTE] Private beta: dev releases only. We currently publish only PEP 440 pre-release (dev) versions and no stable version yet. uv skips pre-releases by default unless they are allowed explicitly:

uv tool install --prerelease allow managed-deepagents

This package requires Python 3.9 or newer. Each platform wheel bundles the prebuilt mda binary for its OS and CPU architecture and exposes it through the mda console script. This PyPI-installed CLI scaffolds and compiles Python projects only and vendors the Python runtime bundled with this wheel.

To start a new project:

mda init my-agent
cd my-agent
uv sync

mda init can shape the project up front — --instructions "..." (or --instructions-file <path>) writes the system prompt, --model <spec> picks the model, --memory agent opts into deployment-shared durable memory, and --no-sandbox leaves out the sandbox. Every new project includes an identity.py that explicitly selects auth.langsmith_api_key() authentication. Managed Deep Agent evals are Harbor tasks under evals/tasks/. Optionally create a minimal source task under evals/scaffold/ with mda evals init <name>, then package the managed agent and scaffolded tasks with mda evals compile .. Nothing prompts, so these commands also work in coding agents and CI scripts.

mda init my-agent --gateway runs the agent on LangSmith Gateway — a model LangSmith hosts, billed to your workspace's Gateway Credits, authenticated with a LangSmith API key instead of a model provider key of your own. It is mutually exclusive with --model, which names a provider you hold the key for.

Define an Agent

Create an agent.py that defines an agent:

from managed_deepagents import define_deep_agent

# The system prompt comes from instructions.md next to this file.
agent = define_deep_agent(
    name="research-assistant",
    model="openai:gpt-5.5",
    tools=[query_db],
)

define_deep_agent requires a static name (LangGraph assistant id and default LangSmith deployment name) and otherwise accepts the create_deep_agent keyword surface minus the managed keys: backend, store, checkpointer, memory, skills, and system_prompt. Those are provided by the managed runtime when your agent is deployed. Write the system prompt in instructions.md next to agent.py; the CLI embeds it at deploy time.

To authenticate SDK and API requests with a LangSmith workspace key while retaining MDA's thread and store authorization, declare it explicitly:

identity = define_identity(auth=auth.langsmith_api_key())

Clients send the key as x-api-key. LangSmith Cloud supplies the verification endpoint and tenant configuration; do not add those platform-owned values to the project .env.

On deploy, Context Hub stores harness files (instructions.md, skills/**). A root memory.py declaring define_memory(scope="agent") additionally mounts one deployment-shared memory tree at /memories/agent/. /memories/agent/AGENTS.md is injected every turn; other files are read on demand. Deploy never overwrites existing memories. Memory is independent of identity, and a project without memory.py mounts no durable memory.

Project Shape

my-agent/
  agent.py              # named `agent` variable
  identity.py           # managed authentication (included by `mda init`)
  instructions.md       # managed system prompt
  pyproject.toml
  .env                  # local deploy secrets, never committed
  schedules/            # optional managed cron schedules
  tools/                # optional custom tools
  middleware/           # optional middleware
  skills/               # optional skills synced to Context Hub
  sandbox/              # LangSmith sandbox (`mda init` includes this; delete to opt out)
  connectors/mcp.py     # optional MCP server declaration

The CLI copies your project files into the managed build and generates the entry module that connects your definition to the hosted runtime.

The agent entry must live at the project root as agent.py.

Define a Schedule

Create one file per schedule under schedules/ and define a named schedule:

# schedules/daily_digest.py
from managed_deepagents import define_schedule

schedule = define_schedule(
    cron="0 8 * * 1-5",
    timezone="America/Los_Angeles",
    prompt="Write the daily digest.",
)

mda deploy reconciles schedules as LangSmith cron jobs after the deployment is live. Declarations must be statically serializable literals or top-level constants; prompt schedules become user-message input, and stateless runs clean up their temporary thread after completion.

Sandbox

mda init scaffolds sandbox/__init__.py with a LangSmith sandbox. MDA only enables the sandbox when that declaration is present — delete sandbox/ to opt out:

from managed_deepagents import define_sandbox

sandbox = define_sandbox(
    scope="thread",
    idle_ttl_seconds=600,
)

If sandbox/setup.sh exists, MDA embeds it and runs it once when the sandbox is first provisioned. MDA owns sandbox naming, image/snapshot selection, reuse, and lifecycle.

MCP Connectors

Add connectors/mcp.py to attach MCP servers. The file must define a module-level connector. By default, MDA exposes every tool loaded from each declared server. Supply your own auth via static headers when the server requires credentials:

from managed_deepagents import connectors

connector = connectors.mcp(
    mcp_servers={
        "langchainDocs": {
            "transport": "http",
            "url": "https://docs.langchain.com/mcp",
            "include_tools": ["search", "fetch"],
        },
    },
)

Use include_tools or exclude_tools inside a server config to select a subset. Tool names are raw MCP tool names before the managed {server}__ prefix is applied, so "include_tools": ["search"] on server langchainDocs exposes langchainDocs__search when prefixing is enabled.

CLI

Create a new project:

mda init my-agent

Build locally:

mda build ./my-agent

Run on the local LangGraph dev server:

mda dev ./my-agent

mda dev requires uv on PATH, but it resolves the local LangGraph dev server automatically; you do not need to install a global langgraph command.

Deploy to LangSmith:

mda deploy ./my-agent

The generated build is written to <root>/.mda/build by default.

Common deploy options:

mda deploy ./my-agent --name my-agent-dev --deployment-type dev
mda deploy ./my-agent --workspace-id "$LANGSMITH_WORKSPACE_ID"
mda deploy ./my-agent --no-wait

Read the deployed agent's server logs:

mda logs ./my-agent
mda logs ./my-agent --lines 200 --level error
mda logs ./my-agent > agent.log

In a terminal mda logs streams new output until you press Ctrl-C. When the output is piped or redirected it prints the most recent lines (1000 by default) and exits.

Tear it down again:

mda delete ./my-agent

mda delete (alias mda destroy) removes the LangSmith deployment, the tracing project created alongside it, the deployment's Context Hub repo including the per-user memory repos beneath it, and the managed sandboxes the deployment created. It asks for confirmation first; pass --yes to skip the prompt in scripts. Agent memory and thread history are not recoverable afterwards.

Sandboxes are matched by name: the runtime names each one {deployment}--{digest}, which also lets a restarted deployment re-adopt its existing sandbox instead of stranding it. The digest covers the sandbox scope plus how it was provisioned, so editing setup.sh or switching snapshots gives the next run a fresh sandbox instead of one built from the previous recipe. Sandboxes created before this behavior existed are unnamed and are left to LangSmith's idle-stop and retention window.

Before deploying, make sure your model provider key such as OPENAI_API_KEY or ANTHROPIC_API_KEY is available in the project .env, the process environment, or LangSmith workspace secrets. For LangSmith itself, set LANGSMITH_API_KEY the same way, or run interactively and press Enter at the prompt to sign in with your browser (the CLI creates a key and writes it to .env). Use LANGSMITH_WORKSPACE_ID or --workspace-id when your credentials require a workspace selection.