@LangChain: You can add schedules to Managed Deep Agents. Add a schedule file to schedules/ with a cron expression + a prompt Deplo…
Summary
LangChain introduces scheduling capabilities for Managed Deep Agents, enabling automated execution of AI agents using cron expressions and prompts.
View Cached Full Text
Cached at: 09/23/26, 10:14 PM
You can add schedules to Managed Deep Agents. Add a schedule file to schedules/ with a cron expression + a prompt Deploy …And the agent runs itself Docs: https://docs.langchain.com/langsmith/python/managed-deep-agents-schedules… Demo from @ndrezn
Add schedules to Managed Deep Agents - Docs by LangChain
Source: https://docs.langchain.com/langsmith/python/managed-deep-agents-schedules
Managed Deep Agents can run agents on a cron schedule. When you deploy the project,mda deployprovisions each schedule as a LangSmith cron after the deployment is live.
Project structure
Schedule declarations live in the project-levelschedules/directory, with one schedule per file:
Add a schedule
The file name becomes the managed schedule name.The schedule module must define a namedscheduledeclaration.
schedules/daily_digest.py
Configure schedule input
Each schedule must define exactly one of:
prompt: A natural-language prompt. Managed Deep Agents converts it to a user message when the cron fires.input: A structured LangGraph input object. Use this when you need to pass custom graph input instead of a single prompt.
schedules/nightly_sweep.py
cronmust be a standard five-field cron expression: minute, hour, day of month, month, and day of week. Iftimezoneis omitted, LangSmith crons use UTC.
Choose thread behavior
Schedules use ephemeral threads by default. Managed Deep Agents creates a fresh thread for each run and asks LangSmith to delete that temporary thread after the run completes.Use a persistent thread only when scheduled runs should accumulate durable thread state across invocations.A persistent thread ID must be the UUID of a thread that already exists on the deployment. Managed Deep Agents passes the value straight to the Agent Server, which rejects any ID that is not a UUID.
Create the persistent thread
Create the thread before you deploy the schedule. Open the deployment inLangSmith Studio, create a new thread, and copy its thread ID.mda deployprints the deployment URL, and LangSmith lists it on the deployment page.Neither Managed Deep Agents nor themdaCLI creates the thread, so this is a one-time manual step for each persistent schedule.
Declare the persistent schedule
Use that thread ID in the schedule declaration.
schedules/nightly_memory.py
Deliver results to Slack
Setdeliver\_toto post the final response through a configuredSlack channel.Use a Slack channel ID because scheduled runs have no originating thread.
schedules/monday_greeting.py
The Slack bot must have access to the destination.
Use static declarations
Schedule declarations are extracted at compile time. Keep schedule configuration statically serializable:
- Use literals, lists, dictionaries, and references to top-level literal constants.
- Do not read environment variables, call functions, use
\*\*kwargs, or compute schedule values dynamically. - Put dynamic behavior in the agent, tools, middleware, or runtime context instead.
Test schedules locally
mda devcompilesschedules/and lists each schedule in its startup banner, but it does not provision them. Schedules run only on a deployment.
To test the agent behavior a schedule triggers, send the schedule’spromptorinputto the agent directly in local Studio. To test the schedule itself, deploy the project to a development deployment.
Deploy schedules
Test the project locally withmda dev, then deploy it withmda deploy. Open deployment traces in LangSmith to inspect model calls, tool calls, errors, and latency.When the deployment reachesDEPLOYED,mda deploysearches for existing Managed Deep Agents-owned cron jobs on the deployed Agent Server, deletes them, and creates cron jobs for the currentschedules/declarations. Removing a local schedule file and redeploying removes the corresponding managed cron.
Troubleshoot schedules
must export a named schedule declaration: Define a top-levelschedulein each file inschedules/.must define exactly one of prompt or input: Add eitherpromptorinput, but not both.cron must be a standard 5\-field expression: Use five cron fields, not seconds-based cron syntax.schedule is not static: Replace computed values with literals or top-level literal constants.failed to create cron for schedule: Open the deployment URL in LangSmith and confirm the deployed Agent Server is healthy.Invalid thread ID: must be a UUID \(HTTP 422\): A persistent schedule declares a thread ID that is not a UUID. Replace it with the UUID of an existing thread. SeeChoose thread behavior.- A schedule does not run during local development:
mda devdoes not provision schedules. SeeTest schedules locally.
Next steps
Similar Articles
@LangChain: Managed Deep Agents v0.9 is here! Schedules SDK: Agents create their own reminders, follow-ups, and recurring tasks mid…
LangChain announces Managed Deep Agents v0.9, adding a Schedules SDK that lets agents create their own reminders and recurring tasks, per-run configuration for models/skills/MCP servers/sandboxes, and Slack Reactions to signal run starts.
@ndrezn: nearly every proactive agent you've seen is a cron job with a prompt memory that "dreams" overnight, agents that ping y…
LangChain explains how nearly all proactive, always-on agents are built on cron-style scheduled runs—covering memory "dreaming," proactivity, triage, follow-ups, and orchestration—and shows how to build them with Managed Deep Agents on LangSmith, where over 70% of runs are already scheduled rather than user-initiated.
@LangChain: You’re an mda init and mda deploy away from distributing a proactive agent in your org. Since launching dynamic schedul…
LangChain reports that since introducing dynamic scheduling for Managed Deep Agents in September, over 70% of MDA runs on LangSmith come from scheduled runs rather than manual prompts, framing proactive agents as easily deployable via `mda init` and `mda deploy`.
@LangChain: Everything you need to know about Managed Deep Agents:
LangChain announces Managed Deep Agents in private beta, a hosted API-first runtime for building, running, and operating deep agents in production, leveraging the open-source Deep Agents harness and integrating with LangSmith for durable execution, streaming, and human-in-the-loop workflows.
@LangChain: PSA: You can connect a Managed Deep Agent to Slack. Mentions, DMs, and thread replies all start a run. Your agent posts…
LangChain announces that users can connect a Managed Deep Agent to Slack, allowing mentions, DMs, and thread replies to trigger agent runs with responses posted in the same conversation.