Understand this story
On the evening of October 31, 2026, the DeepSeek Harness team released "Super-agent Design Exploration: Toward More Autonomous Agent Systems," proposing the Super Agent concept and open-sourcing the prototype dsh-superagent-prototype on GitHub. According to Sina Tech, the architecture consists of five modules: a two-layer Router/Worker, event-driven operation, asynchronous multi-channel interaction, storage-compute separation, and an external Watchdog. The goal is to enable Agents to stay continuously online, run multiple tasks in parallel, and proactively respond to environmental events; in preliminary experiments, the p50 task completion time was 12.4 minutes and the p95 was 15.7 minutes, and Router batching reduced the inference time allocated to a single decision from 6.4 seconds to 1.6 seconds. TMTPost·AGI reported that the article splits tasks using two roles, Router and Worker, triggers work from activity such as group chats, emails, and code checks through two filters—rules plus models—and emphasizes notifying only when results are usable or a human decision is needed. The team stated that the prototype is only a research prototype and not an official DeepSeek product; it currently mainly targets a GitHub repository and can also connect to Feishu, and some designs have not yet been implemented.
AI-generated from public reports
Latest development · Report
Digital Life Kazik shares a year of practice building a self-hosted Agent automation systemDevelopments
1 confirmed developments
DeepSeek Harness team shares exploration of Super-agent design
Source:TMTPost · AGI
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 2
Kazik
Digital Life Kazik shares a year of practice building a self-hosted Agent automation system
Digital Life Kazik shared the Agent automation system he spent nearly a year building; the core is shifting task triggers from people to events, with people only handling the parts that require decisions. The system includes an automation middle platform that uniformly receives and distributes events, an execution layer that assigns models such as Claude Opus 5.5 and GPT-6.1 Sol/Terra by task, a self-built five-step CI pipeline on an M4 Mac Mini, and an approval step completed via Feishu cards. The author says that in the past two weeks, four projects reviewed a total of 298 merge requests, Codex reported 333 issues, of which 306 were fixed before merging.
East Money
DeepSeek Harness Team Releases Super-agent Architecture
On the evening of October 31, the DeepSeek Harness team published a technical article titled "Super-agent Design Exploration: Toward a More Autonomous Agent System," proposing to rebuild the Agent system around five major modules—a two-layer architecture, event-driven design, asynchronous multi-channel interaction, storage-compute separation, and an external Watchdog—summarized as four characteristics: continuous, efficient, proactive, and persistent. In synthetic load tests with about 300 messages per run, of which about 20 were work requests, the Super-agent's median and 95th-percentile task completion times were 12.4 minutes and 15.7 minutes, respectively, while the single-session approach under neutral prompts had a 95th-percentile time of 29.5 minutes. In the Router-layer stress test, the event arrival rate rose from about 1 per minute to about 40 per minute, and the inference time apportioned to a single decision dropped from 6.4 seconds to 1.6 seconds.
East MoneyRelated discussion
East Money joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
X.PIN
DeepSeek Harness team plans to restructure agent sessions with Router and Worker
DeepSeek's Harness team proposes replacing the existing agent session model with a Router responsible for conversing with users and a Worker responsible for executing tasks in isolated sessions. The team says that a person can only track 3-5 active sessions at a time, so what needs to change is the session as a unit, not model capability; persistent agents need threads.
Nov 1
A Yi AI Notes
DeepSeek publishes a 14,000-word technical deep dive on Super-agent
DeepSeek publishes a 14,000-word technical deep dive on Super-agent, pointing out that the real bottleneck to Agent deployment is human attention, not model capability. Its architecture consists of a lightweight Router that assigns work, context-isolated Workers that execute single-point tasks, a subscription mechanism that reaches the responsible person directly, and an independent Watchdog for inspection patrols, reducing task completion time from 29.5 minutes in a single session to 12.4 minutes. The article also states that replies within the chat software are limited to 100 characters.
Alibaba Cloud Developer Community AI
DeepSeek Open-Sources Super-agent Prototype: How Should Testers Respond?
DeepSeek's Harness team published a technical article titled "Super-agent Design Exploration: Toward a More Autonomous Agent System" and open-sourced a prototype plugin that lets AI autonomously discover issues in group chats, reproduce them in the background, and submit Pull Requests. The article introduces four capabilities: continuous cross-session interaction, multi-task parallelism, proactive event response, and long-term online presence. It adopts a two-layer Router + Worker architecture with a Watchdog supervision mechanism. DeepSeek's internal testing shows that in a simulated message stream of about 300 messages, Super-agent's p95 task completion time was 15.7 minutes, compared with 29.5 minutes for the traditional single-session approach, but cache-read Token volume rose from 23.23 million to 31.77 million.
English text is being prepared.Related discussion
English text is being prepared. joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
English text is being prepared.Related discussion
English text is being prepared. joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
Tianyi CuiRelated discussion
Tianyi Cui joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
TMTPost · AGISelected
The DeepSeek Harness team proposes the Super Agent design and open-sources a prototype
On the evening of October 31, the DeepSeek Harness team released "Super-agent Design Exploration: Toward a More Autonomous Agent System," proposing the Super Agent concept and open-sourcing the prototype dsh-superagent-prototype on GitHub. The article splits tasks between two roles, Router and Worker, triggers work from activity such as group chats, emails, and code checks through two filters—rules plus a model—and emphasizes notifying only when results are usable or a human decision is needed. The team says the prototype is only a research prototype, not a formal product; it currently mainly targets one GitHub repository and can also connect to Feishu, and some designs have not yet been implemented.
TMTPost · AGIRelated discussion
TMTPost · AGI joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
MarkTechPostRelated discussion
MarkTechPost joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
Berry XiaRelated discussion
Berry Xia joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
Phoenix Technology
DeepSeek Harness team proposes Super Agent design and open-sources prototype
On the evening of October 31, the DeepSeek Harness team released "Super-agent Design Exploration: Toward a More Autonomous Agent System," proposing the Super Agent concept and open-sourcing the prototype dsh-superagent-prototype on GitHub. In this design, a Router is responsible for task dispatch, while multiple Workers execute in parallel within isolated contexts; triggers such as group chats, emails, and code checks are identified through a two-stage filter of rules plus a model, and the design emphasizes notifying only when results are available or a human decision is needed. The team says the prototype currently mainly targets a single GitHub repository, with optional Feishu integration; the asynchronous task mechanism has been implemented in DSH but is not yet enabled by default, and some parts of the design have not yet been implemented.
Yicai
AI Weekly: iPhone China AI to Open Testing Next Week; DeepSeek's 100-Billion-Yuan Financing Details Revealed
Apple announced that the first batch of Apple Intelligence features will open for testing in China on November 4 with the iOS developer beta, and will officially launch in December with iOS 27.3. It is also partnering with Alibaba to integrate Qwen into Siri, Writing Tools, and Visual Intelligence. DeepSeek completed its second round of financing this month, raising approximately 105.5 billion yuan, with a post-investment valuation of 615.5 billion yuan; Tencent and CATL remain the two largest investors. Anthropic has submitted filings to list on Nasdaq under the ticker ANT, with market valuation expectations of 1.5 trillion to 2 trillion US dollars.
East Money
AI Weekly: iPhone China AI Opens Testing on November 4, DeepSeek's Hundred-Billion Funding Details Revealed
Apple announced that the first batch of Apple Intelligence features will open for testing in China on November 4 with the iOS developer Beta, and will officially launch in December with iOS 27.3. It is also partnering with Alibaba to integrate Qwen into Siri, Writing Tools, and Visual Intelligence. DeepSeek completed its second round of funding of approximately RMB 105.5 billion this month, with a post-investment valuation of RMB 615.5 billion; Tencent and CATL remain the two largest investors. Anthropic has filed with the SEC and plans to list on Nasdaq under the ticker "ANT".
East Money
DeepSeek Harness team publishes exploration of Super-agent architecture design
The DeepSeek Harness team published a technical article titled "Super-agent Design Exploration: Toward a More Autonomous Agent System" on its official WeChat account, proposing a dual-layer architecture, event-driven design, asynchronous multi-channel interaction, separation of storage and compute, and an external Watchdog that takes over session management and context synchronization, shifting the multitasking ceiling from human attention to the system layer. The team says existing Agents operate as single Sessions that are independent of one another, and a person can typically follow only 3 to 5 active Sessions at the same time. Based on this approach and DeepSeek Harness, an internal Super-agent instance has been running continuously, participating in issue reproduction and fixes, code review responses, and iteration on Harness itself; the team emphasizes that this is an engineering-practice-oriented discussion of architectural ideas, not an introduction to existing product features.
Sina TechSelected
DeepSeek open-sources Super-agent architecture prototype, unveils five-module design
The DeepSeek Harness team has unveiled the underlying architecture of Super-agent, an autonomous Agent exploration project, and open-sourced the prototype dsh-superagent-prototype on GitHub. The architecture consists of five modules: a two-tier Router/Worker, event-driven design, asynchronous multi-channel interaction, storage-compute separation, and an external Watchdog. Its goal is to enable the Agent to stay continuously online, run multiple tasks in parallel, and proactively respond to environmental events. In preliminary experiments, the p50 task completion time was 12.4 minutes and the p95 was 15.7 minutes, and Router batching reduced the inference time amortized per individual decision from 6.4 seconds to 1.6 seconds. The team stated that the experiments were based on synthetic workloads and a simplified environment, and that this plugin is only a prototype for validating the technical approach, not an official DeepSeek product.
Sina TechRelated discussion
Sina Tech joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
IT Home · Artificial Intelligence
DeepSeek proposes Super-agent architecture: a two-tier design with Router and Worker
The official DeepSeek Harness team released "Super-agent Design Exploration," proposing a two-tier architecture composed of a Router layer and a Worker layer. The Router handles human interaction, maintains global context, and dispatches tasks, while Workers execute concurrently in isolated sessions. The team says the architecture has four characteristics—continuous, efficient, proactive, and persistent—and incorporates mechanisms such as event-driven operation, Inline & Spill, chain recall, speculative execution, and Watchdog. In preliminary experiments, the p50 and p95 task completion times for Super-agent were 12.4 minutes and 15.7 minutes, respectively, while the p95 for a single-session approach under neutral prompts was 29.5 minutes; the team notes that the experiments were based on synthetic workloads and a simplified environment. The article also provides a prototype implementation named dsh-superagent-prototype and makes clear that it is not an official DeepSeek product.
IT Home · AIRelated discussion
IT Home · AI joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
Zhidx
DeepSeek Harness unveils Super-agent architecture, using five major modules to take over session scheduling
The DeepSeek Harness team has unveiled the underlying architecture of Super-agent, an autonomous Agent exploration project, bringing classic computer systems ideas such as scheduling, caching, compute-storage separation, and Watchdog into Agent design. The architecture consists of five modules: a dual-layer architecture, event-driven design, asynchronous multi-channel interaction, compute-storage separation, and an external Watchdog. Its goal is to enable the Agent to stay continuously online, advance multiple tasks in parallel, and proactively respond to environmental events. In preliminary experiments, Super-agent's p50 and p95 task completion times were 12.4 minutes and 15.7 minutes, respectively. After removing subscription passthrough, the number of Router decisions increased by about 30%, and read-cache tokens rose from 23.23 million to 31.77 million. The team said the experiments were based on synthetic workloads and a simplified environment, and that this plugin is only a prototype for validating the technical approach, not an official DeepSeek product.
Sina Tech
DeepSeek Harness Team Releases Super-agent Architecture
The DeepSeek Harness team published a technical article on its official WeChat account, proposing an Agent form and system architecture called Super-agent. Through designs including a dual-layer architecture, event-driven operation, asynchronous multi-channel interaction, storage-compute separation, and an external Watchdog, it enables the Agent to maintain context across sessions and advance multiple tasks in parallel. The architecture has four basic characteristics: continuous, efficient, proactive, and persistent, and users can interact with the same Agent through interfaces such as IM chats and GitHub comments. The team said that internal instances built on this approach are already running continuously, participating in issue reproduction and fixing, code review responses, and the iteration of DeepSeek Harness itself.
Oct 31
East Money
The DeepSeek Harness team publishes a technical article on Super-agent design exploration
The DeepSeek Harness team published "Super-agent Design Exploration: Toward a More Autonomous Agent System" on its official WeChat account, proposing five modules—a dual-layer architecture, event-driven design, asynchronous multi-channel interaction, storage-compute separation, and an external Watchdog—to support an Agent that is continuously online and responds concurrently. The team says an internal Super-agent instance built on this approach is already running continuously, participating in issue reproduction and fixes, code review responses, and iteration of DeepSeek Harness itself. In preliminary ablation experiments, the Super-agent's task completion time p50 and p95 were 12.4 minutes and 15.7 minutes, respectively, outperforming the single-Session approach with Subagent tools.
Yicai
DeepSeek Releases Super-agent Architecture; Internal Instance Already Participates in Code Review and Fixes
The DeepSeek Harness team published a technical article titled "Super-agent Design Exploration: Toward a More Autonomous Agent System," proposing a dual-layer architecture, event-driven design, asynchronous multi-channel interaction, storage-compute separation, and an external Watchdog to break through the scale limits of multitasking and team collaboration. The team said that an internal Super-agent instance built on this approach is already running continuously, routinely participating in issue reproduction and fixes, code review responses, and DeepSeek Harness's own iteration; when members report software issues in group chats, even if no one @s the AI, it can identify potential defects and attempt in the background to reproduce them and find fixes.
YicaiRelated discussion
Yicai joined this discussion
This post contributes to heat; it is not a formal report or confirmed fact.
Shao Meng
The DeepSeek Harness team shares the Super-agent architecture design: using...
The DeepSeek Harness team shares its Super-agent design exploration, arguing that single-task Agents are already strong enough and that the bottleneck has become the roughly 3–5 active Sessions a person can monitor at once; the Agent is therefore made into an entity that is continuous across channels, parallelizable, event-driven, and persistently online in the cloud. The proposal includes a two-layer architecture of a resident Router plus Workers with isolated contexts, two-level Trigger filtering with direct Thread pass-through, asynchronous ReAct, separation of the cloud Loop from the device-side daemon, and an external Watchdog. In synthetic load tests, task completion time p50/p95 was about 12.4/15.7 minutes, better than the roughly 29.5-minute p95 of a single Session plus Subagent, at the cost of higher read-cache tokens.
QbitAI
DeepSeek Unveils Super-agent Architecture: Two-Tier Router and Worker Design
DeepSeek has published an engineering-oriented architecture article titled "Super-agent: Toward More Autonomous Agent Systems," proposing that an upper-layer Router is responsible for receiving messages, assigning tasks, and maintaining global context, while lower-layer Workers execute specific tasks in parallel. In a simulated environment with approximately 300 messages and 20 work requests, repeated tests 5 times showed that the Super-agent's median task completion time was 12.4 minutes and its 95th percentile was 15.7 minutes, whereas the traditional single-Session approach had a 95th percentile of 29.5 minutes under neutral prompts. The article also introduces two-level event filtering, an asynchronous ReAct Loop (already implemented in DeepSeek Harness but not enabled by default), separation of computation and state from the execution environment, and a Watchdog supervision mechanism.
Phoenix Technology
DeepSeek proposes Super-agent architecture: a two-tier design of Router and Worker
The official DeepSeek Harness team released "Super-agent Design Exploration," proposing a two-tier architecture in which a Router layer uniformly handles human interaction and task dispatch while multiple Workers execute concurrently in isolated sessions, and summarizing four characteristics: continuous, efficient, proactive, and persistent. Experiments show that Super-agent's task completion time p50 and p95 were 12.4 minutes and 15.7 minutes respectively, while the single-Session approach under neutral prompts had a p95 of 29.5 minutes; the team noted that the experiments were based on synthetic workloads and a simplified environment, mainly evaluating scheduling and collaboration capabilities. The team also provided a prototype implementation named dsh-superagent-prototype, and made clear that it is intended to validate the technical approach and is not an official DeepSeek product.
Heat trend for this story
- Current
- 88
- Period peak
- 112 Nov 2 01:00
- 24-hour change
- +24%
Move the pointer or select the chart to inspect hourly heat. Use the left and right arrow keys with a keyboard.