Sqwibbl: The First Windows-Native Computer Use Agent Powered by the PLM Network

(YourDigitalWall Editorial):- Durham, North Carolina Jul 7, 2026 (Issuewire.com) – Professional Language Models, Inc. (PLM, Inc.) today announced the general availability of Sqwibbl 1.0.0, a groundbreaking Windows native personal AI agent. Unlike conventional AI assistants that live in browser tabs and rely entirely on cloud connectivity, Sqwibbl operates directly on the user’s desktop, featuring its own onboard Large Language Model (LLM) and native integration with the decentralized expert knowledge infrastructure of the PLM Network.

A Native Windows Agent, Not a Browser Tab

Sqwibbl is designed to live where actual work happens. Installing directly as a local Windows application, Sqwibbl operates from the system tray and serves a local control panel, ensuring low-latency performance and robust privacy.

By default, Sqwibbl’s local control surface binds exclusively to the user’s machine, keeping durable local memory, chat history, and contextual data secure. It can read, write, and create files, automate Windows desktop applications via UI Automation, interact with Google Workspace, and generate Microsoft Office documents natively–without requiring Office to be installed.

“The product goal for Sqwibbl is simple: Make powerful agent capability feel safe, visible, useful, and normal on a Windows PC. We didn’t want to build another chat window. We built a capable assistant that actually understands your desktop environment and can take action on your behalf, grounded by the world’s best human expertise.” — Bradley P. Jones, Co-Founder & CEO, Professional Language Models, Inc.

The Onboard Brain: Local AI Meets Collaborative Superintelligence

Sqwibbl distinguishes itself through its sophisticated Brain Router architecture. Instead of assuming one model is right for everything, Sqwibbl dynamically routes requests to the most capable engine.

At the core of this system is Sqwibbl’s onboard local LLM. Upon first launch, Sqwibbl evaluates the user’s hardware and downloads the optimal local model tier–ranging from CPU-optimized models to the high-performance Power Sqwibbl profile utilizing Gemma 4 for machines with NVIDIA GPUs. This ensures that routine tasks, file operations, and contextual summarization happen instantly and privately on the user’s local hardware.

However, when complex reasoning or highly specialized domain knowledge is required, Sqwibbl seamlessly escalates. It is the first agent natively built to connect to PLMMarket.com and the PLM Network.

Backed by Professional Language Models (PLMs)

Artificial intelligence faces a fundamental limitation: general-purpose models hallucinate when confronted with niche, proprietary, or highly specialized professional queries. Sqwibbl solves this by connecting to Professional Language Models (PLMs).

A PLM is a domain-specific knowledge container created and maintained by a verified human expert. When a Sqwibbl user asks a highly technical, legal, medical, or strategic question, Sqwibbl does not guess. It routes the query to the PLM Network, retrieving verified facts directly from the experts who created them. This architecture–termed Collaborative Superintelligence (CSI)–combines the deep reasoning of frontier models with the grounded, factual truth of human professionals.

Through this integration, Sqwibbl users gain direct access to the world’s most valuable “Tribal Knowledge,” while the human experts who minted those PLMs are compensated via machine speed micropayments for every query their knowledge resolves.

Action Cards: Safety and Visibility First

Because Sqwibbl is capable of powerful local and external actions, safety and user control are paramount. Sqwibbl introduces Action Cards–a visible approval layer for any operation that changes files, alters the desktop, or interacts with external financial or sensitive accounts.

Action Cards present the user with a clear description of what Sqwibbl intends to do, allowing the user to review, edit parameters, approve, or cancel the action before execution. This ensures that the user is always in control and can clearly see the evidence and reasoning behind every agentic action.

Availability

Sqwibbl 1.0.0 is available now. Users can download the stable Windows installer and register their agent identity to connect to the PLM Network.

For developers, Sqwibbl offers a Capability SDK to build installable contracts, custom skills, and visual UI descriptors, extending the agent’s power across the Windows ecosystem.

To download Sqwibbl or learn more about the PLM Network, visit sqwibbl.ai and plmmarket.com.

About Professional Language Models, Inc.

Professional Language Models, Inc. is headquartered in Durham, North Carolina. Co-founded by Bradley P. Jones and Philip Bridgeman, PLM, Inc. is the architect of the Decentralized Intelligence Economy and the builder of the PLM Network. By transforming human expertise into owned, monetizable digital property, PLM, Inc. powers Collaborative Superintelligence and delivers the most capable, grounded AI agents in the world.

Media Contact

Bradley P. Jones, Co-Founder & CEO

Philip Bridgeman, Co-Founder & CINO

Professional Language Models, Inc. | Durham, North Carolina

Email: [email protected] | [email protected]

Web: plmmarket.com | sqwibbl.ai