Get the latest tech news

Deep Agents

Using an LLM to call tools in a loop is the simplest form of an agent. This architecture, however, can yield agents that are “shallow” and fail to plan and act over longer, more complex tasks. Applications like “Deep Research”, “Manus”, and “Claude Code” have gotten around this limitation by

Applications like “ Deep Research ”, “ Manus ”, and “ Claude Code ” have gotten around this limitation by implementing a combination of four things: a planning tool, sub agents, access to a file system, and a detailed prompt. In order to make it easier for everyone to build a deep agent for their specific vertical, I hacked on an open source package ( deepagents) over the weekend. A system prompt inspired by Claude Code, but modified to be more general A no-op Todo list planning tool (same as Claude Code) Ability to spawn sub-agents, and specify your own A mocked out “virtual file system” that uses the agents state (a preexisting LangGraph concept)

Get the Android app