Taskyon

Taskyon is a browser-based AI workspace for people investigating technical problems and developing engineering designs. It keeps the work in a task tree so related questions, evidence and decisions stay connected.

• Task trees: Turn conversation messages into task nodes and inspect the sources, assumptions, prompts and outputs associated with the work.
• Web research: Create research branches that use web search, page crawling and browser tools imported through MCP.
• Tool and code execution: Tasks can call tools and functions or run JavaScript/Python-style workflows in sandboxed environments.
• Files and technical output: Attach files and data as context, then present results with mathematical notation, Mermaid diagrams, SVG and HTML widgets.

You can run Taskyon locally in a browser or use the hosted app at taskyon.space. It's open source under the MIT license and can be embedded in other web pages or local workflows.

Workflow data and task state stay local unless you share them or a task requires an external service. Model requests can go to OpenAI-compatible hosted or self-hosted endpoints, so the endpoint you choose determines where those requests go. Web research can also send requests to external services.