Qwen Open-Sources MyContext, Marking the Arrival of the Enterprise Context Era

Building a New Context Infrastructure for Agents

Qwen Office has officially open-sourced a new context infrastructure called "MyContext."

It consolidates fragmented personal work data from various sources and formats into dedicated work profiles that are easily understood by Agents. This enables Agents to truly understand users and real-world business workflows, achieving efficient human-AI collaboration in decision-making and execution, significantly enhancing their performance in office scenarios.

Just one week after launch, MyContext has already gained over 1,000 stars on GitHub, with growing popularity.

Enabling Agents to Truly Understand Users and Real Business Workflows

Why Context Infrastructure Is Needed

As model capabilities rapidly improve and Agent Harness architectures evolve, general-purpose Agents are becoming increasingly capable in office tasks, significantly boosting individual efficiency. However, due to the lack of real-world contextual information, these Agents struggle to integrate into actual business processes and thus fail to effectively enhance organizational productivity. According to MIT's NANDA report, about 95% of enterprise-level generative AI pilot projects fail to deliver value, primarily due to the absence of proper data infrastructure, preventing AI systems from integrating into existing workflows.

A Continuously Updated Work Profile

Unlike traditional knowledge bases or vector databases, MyContext operates locally on user devices. It automatically aggregates multi-source data from instant messaging, documents, meeting notes, collaboration platforms, and both online and offline activities, building a continuously updated personal work profile. This profile includes job responsibilities, collaboration relationships, behavioral habits, and discussion outcomes, enabling Agents to quickly query and reason. Each piece of information is traceable to its original source, clearly labeled with origin, timestamp, and content, ensuring transparency and effectively mitigating AI hallucinations, greatly improving task accuracy.

The Technical Innovation Behind MyContext

Qwen Office employs innovative algorithms and engineering designs when handling critical tasks such as sequential data and conflicting data. For example, when faced with contradictory messages in instant messaging (IM), traditional methods typically adopt either "take the latest" or "take the highest confidence" strategies—both of which can lead to information loss in Agent scenarios. MyContext uses semantic analysis to determine whether conflicting information should be merged or overwritten; if semantic analysis cannot make a determination, it defers to user confirmation. Once confirmed, this information cannot be overwritten by the model, thereby greatly ensuring data validity and consistency.

The Era of Enterprise Context Is Approaching

MyContext supports integration with mainstream personal IM platforms such as DingTalk and Feishu. After importing their data, users can build personalized Agent assistants that automatically reply to work messages in a manner consistent with their own tone and style, based on summarized personal knowledge. In the future, the team plans to release an enterprise version of the context infrastructure, fully supporting major enterprise systems like Salesforce and SAP, as well as local workflows, helping organizations achieve a leap in productivity at scale.

We dedicated to serving clients with professional DingTalk solutions. If you'd like to learn more about DingTalk platform applications, feel free to contact our online customer service or email at This email address is being protected from spambots. You need JavaScript enabled to view it.. With a skilled development and operations team and extensive market experience, we’re ready to deliver expert DingTalk services and solutions tailored to your needs!