# How to implement Dynamic Runtime Subagents in DeepAgents + LangGraph **URL:** \ **Category:** Deep Agents\ **Tags:** intro-to-langgraph, python-help\ **Created:** [September 25, 2026, 6:40am UTC](https://forum.langchain.com/t/how-to-implement-dynamic-runtime-subagents-in-deepagents-langgraph/4607 "2026-09-25T06:40:19Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![mrsknetwork](https://avatars.discourse-cdn.com/v4/letter/m/bcef8e/32.png) [@mrsknetwork](https://forum.langchain.com/u/mrsknetwork)\ **Post date:** [September 25, 2026, 6:40am UTC](https://forum.langchain.com/t/how-to-implement-dynamic-runtime-subagents-in-deepagents-langgraph/4607/1 "2026-09-25T06:40:19Z")
Hi everyone, I’m trying to understand the best way to implement **dynamically generated subagents at runtime.** My goal is slightly different from the standard DeepAgents subagent pattern. Instead of predefining all subagents, I want the orchestrator to: 1. Analyze the incoming task. 2. Generate a specialist agent definition at runtime. 3. Allocate a custom system prompt, tools, permissions, workspace/file access rules, and context. 4. Execute the specialist. 5. Verify the output using a reviewer/critic agent. 6. Return the result to the orchestrator. 7. Dispose of the temporary specialist and continue the workflow. ### Conceptually: Orchestrator ↓ Agent Architect ↓ Runtime Agent Factory ↓ Specialist Agent ↓ Verifier ↓ Aggregator From my research, it seems DeepAgents supports custom “ **CompiledSubAgent** ” runnables, and LangGraph provides dynamic worker execution through “ **Send** ”. This makes me wonder whether a runtime-created LangGraph workflow or DeepAgent instance could be treated as a temporary specialist worker. ### My questions: - Is this architecture aligned with how DeepAgents and LangGraph are intended to be used? - Has anyone implemented runtime-generated specialists or agent factories in production? - Would you recommend using LangGraph as the orchestration layer and DeepAgents as workers, or keeping everything inside DeepAgents? - Are there existing examples of combining “CompiledSubAgent”, “Send”, and dynamic tool allocation for this pattern? ---
**Author:** ![capitaharlock](https://avatars.discourse-cdn.com/v4/letter/c/43a26b/32.png) [@capitaharlock](https://forum.langchain.com/u/capitaharlock)\ **Post date:** [September 25, 2026, 7:26am UTC](https://forum.langchain.com/t/how-to-implement-dynamic-runtime-subagents-in-deepagents-langgraph/4607/2 "2026-09-25T07:26:03Z")
In your Runtime Agent Factory, who authorizes the specialist’s tools and workspace/file access: the generated definition itself, or runtime checks independent of it? Putting permission rules in the system prompt isn’t the same as enforcing them. ---
**Author:** ![mrsknetwork](https://avatars.discourse-cdn.com/v4/letter/m/bcef8e/32.png) [@mrsknetwork](https://forum.langchain.com/u/mrsknetwork)\ **Post date:** [September 25, 2026, 8:15am UTC](https://forum.langchain.com/t/how-to-implement-dynamic-runtime-subagents-in-deepagents-langgraph/4607/3 "2026-09-25T08:15:15Z")
Current **workspace-level** access grants permissions to all sub-agents accessible by the main orchestration agent. ---
**Author:** ![keenborder786](https://yyz1.discourse-cdn.com/flex007/user_avatar/forum.langchain.com/keenborder786/32/1757_2.png) [@keenborder786](https://forum.langchain.com/u/keenborder786)\ **Post date:** [September 25, 2026, 7:44pm UTC](https://forum.langchain.com/t/how-to-implement-dynamic-runtime-subagents-in-deepagents-langgraph/4607/4 "2026-09-25T19:44:35Z")
Hello @mrsknetwork welcome to Langchain community, This is a reasonable architecture, but separate **runtime-generated plans** from **runtime-generated permissions**. I would generally use LangGraph for orchestration and DeepAgents as workers when specialists need filesystem access, long-running planning, or their own subagents. `Send` is useful for dynamic fan-out, while `CompiledSubAgent` can represent a worker but neither provides authorization or sandboxing. The agent architect should return a validated declarative spec containing things like the model, tools, workspace, and budget. A separate policy layer must authorize that spec before execution: ```plaintext effective permissions = caller permissions ∩ environment policy ∩ approved capabilities ``` Never let the generated prompt grant permissions or load arbitrary tools. If all subagents inherit workspace-level access, they share that authority; narrower access must be enforced by tool wrappers, scoped filesystems, credentials, or isolated processes. For verification, use deterministic checks followed by a reviewer with limited permissions. The reviewer should return a structured approval, retry, or rejection decision. For simple model routing, you likely do not need to compile a new graph per request. Classify once, store the route in state, and select from a static allowlist of models. This is easier to observe, secure, and maintain. ---
**Author:** ![keenborder786](https://yyz1.discourse-cdn.com/flex007/user_avatar/forum.langchain.com/keenborder786/32/1757_2.png) [@keenborder786](https://forum.langchain.com/u/keenborder786)\ **Post date:** [October 1, 2026, 9:56am UTC](https://forum.langchain.com/t/how-to-implement-dynamic-runtime-subagents-in-deepagents-langgraph/4607/5 "2026-10-01T09:56:20Z")
@mrsknetwork if everything is clear, please mark the answer as solution otherwise let me know if you have any more questions, would love to help you out 🙂 ---
**Author:** ![mrsknetwork](https://avatars.discourse-cdn.com/v4/letter/m/bcef8e/32.png) [@mrsknetwork](https://forum.langchain.com/u/mrsknetwork)\ **Post date:** [October 7, 2026, 4:34pm UTC](https://forum.langchain.com/t/how-to-implement-dynamic-runtime-subagents-in-deepagents-langgraph/4607/6 "2026-10-07T16:34:52Z")
@keenborder786 thanks for your suggestion and solution. I also noticed that Langchain recently released their beta version of the SDK update that has dynamic sub-agents. Which I was looking for. [Dynamic subagents - Docs by LangChain](https://docs.langchain.com/oss/python/deepagents/dynamic-subagents)