DeepSeek releases Super-agent architecture proposal, using a Router to schedule multiple coding agents
On October 31, the DeepSeek Harness team released a technical proposal titled "Exploring Super-agent Design: Towards a More Autonomous Agent System," proposing the use of a central Router to distribute tasks to Worker agents executing in parallel, in order to reduce the burden on developers of managing coding sessions one by one. In the simulated experiments, the 95th percentile task completion time for this architecture was 15.7 minutes, compared with 29.5 minutes for the single-session approach; each run involved about 300 messages and about 20 task requests, each configuration was tested 5 times, and the task workload was synthetic data, with coding tasks simulated by time-consuming tools, so what was measured was scheduling and coordination capability rather than coding capability. DeepSeek also open-sourced the research prototype dsh-superagent-prototype on GitHub under the MIT license, which supports a single repository and can optionally listen for Feishu messages, and stated that the prototype is not an official product and differs from the internal implementation, while also flagging security risks such as prompt injection and credential leakage.
Why read this
AI score 64
DeepSeek has publicly disclosed the design and simulated experiment data of a multi-agent scheduling architecture, along with prototype code under the MIT license, for readers interested in coding agent orchestration to reference.
Topics
Body
Chinese AI company DeepSeek on October 31 published a technical proposal for a ‘super-agent’ architecture designed to coordinate multiple artificial intelligence agents with limited human intervention. The proposal, titled ‘Exploring Super-agent Design: Towards a More Autonomous Agent System’, was published by the DeepSeek Harness team on its official WeChat account.
The article outlines an approach to reduce the time developers spend managing individual AI coding sessions. DeepSeek estimated that a person could typically oversee three to five active agent sessions simultaneously, making human attention a constraint on deploying multiple AI assistants.
The proposed architecture separates coordination from execution. A central agent, called a Router, receives messages and distributes work to independent agents, or Workers, which execute tasks in parallel and report their results. The Router handles user interactions while the Workers perform assigned tasks.
The system is also designed to respond to events such as messages and code updates without requiring explicit instructions for every task. DeepSeek described an example in which an agent detects a software bug reported in a group chat, attempts to reproduce and fix it, and submits a proposed code change for human review.
In simulated experiments, DeepSeek’s super-agent architecture recorded a 95th-percentile task completion time of 15.7 minutes, compared with 29.5 minutes for a single-session approach using a neutral prompt and a subagent tool.
The experiment involved about 300 messages per run, including around 20 work requests, with each configuration tested five times. DeepSeek said the workload was synthetic and coding tasks were simulated using a tool that consumed time. The tests measured scheduling and coordination capabilities, not the system’s ability to write or fix code.
The proposal comes as AI developers expand coding assistants. Anthropic’s Claude Code and OpenAI’s Codex are among the tools designed to handle increasingly complex software development tasks.
DeepSeek has also published an open-source research prototype, dsh-superagent-prototype, on GitHub under the MIT License. The prototype supports a single GitHub repository and can optionally listen to messages on Feishu, a workplace collaboration platform.
The company said the public prototype is not an official product and differs from its internal implementation. Some architectural ideas in the technical article remain future development directions and are not implemented in the released code.
DeepSeek also acknowledged security and privacy risks associated with autonomous agents connected to external services, including prompt injection and potential credential exposure. The company described the public code as a research prototype rather than production-ready software.