Bluesky is expanding beyond its core social networking features with a new AI-driven tool called Attie, designed to help users build customized social media feeds using natural language prompts. The move reflects a broader shift among platforms experimenting with AI assistants, particularly as personalization becomes a central part of how users navigate online content.
Attie is positioned as a way to simplify feed creation, removing the need for technical knowledge or manual configuration. Instead of adjusting filters or writing code, users can describe what they want to see. For example, prompts like “electronic music and experimental sound from people in my network” or “builders working on agent infrastructure” can be used to generate a tailored feed. The system interprets these requests and constructs a feed accordingly, effectively turning feed design into a conversational process.
The assistant is built on the AT Protocol, Bluesky’s open-source framework, which underpins the platform’s decentralized approach. This foundation allows for more flexibility in how content is organized and shared across applications. While Attie currently exists as a separate app, its shared infrastructure with Bluesky suggests potential integration points in the future, either directly within the platform or across other apps that adopt the same protocol.
Bluesky’s leadership has described Attie as an “agentic” tool, meaning it acts on behalf of the user rather than simply responding to queries. In practice, this positions the assistant somewhere between a traditional chatbot and a utility layer that actively shapes the user’s experience. The concept aligns with a growing trend in AI development, where tools are expected not just to answer questions but to perform tasks and adapt interfaces dynamically.
At launch, Attie is being tested through an invite-only beta, with a waitlist open to users who want early access. This limited rollout suggests Bluesky is still refining how the assistant performs in real-world scenarios, particularly in terms of accuracy, relevance, and user control.
The introduction of Attie also highlights how social platforms are rethinking discovery. Instead of relying solely on algorithms that passively recommend content, tools like this allow users to define their own parameters more directly. Whether that approach leads to better engagement or simply adds another layer of complexity remains to be seen.
As AI assistants become more common across apps, Bluesky’s approach focuses on customization and openness rather than tightly controlled ecosystems. The long-term impact will likely depend on how well these tools balance user control with ease of use, and whether they can deliver consistent, meaningful results beyond initial experimentation.
