Show HN: BitFun – An Agentic Development Environment (Rust and TypeScript)(github.com)
github.com
Show HN: BitFun – An Agentic Development Environment (Rust and TypeScript)
https://github.com/GCWing/BitFun
2 コメント
The distinction between "AI as chatbox attached to an editor" and "AI as agent that collaborates across the whole development workflow" is where most of the interesting tooling exploration is happening right now. The chatbox model is familiar and low-risk; the agentic model is more powerful but requires different assumptions about trust, verification, and the division of responsibility.
The Plan Mode / Debug Mode / Review Mode differentiation is thoughtful — it acknowledges that different tasks require different levels of agent autonomy and different kinds of human oversight. Giving the developer explicit control over which mode is active is a meaningful design choice.
The Agile Vibe Coding Manifesto's principle that "humans remain accountable for software systems" is directly relevant here: accountability doesn't mean doing everything yourself, but it does mean maintaining enough understanding and control to be genuinely responsible for the outcomes. Different agent modes with different oversight requirements is one way to structure that.
The "rules to inject repository conventions" feature is particularly interesting for teams: https://agilevibecoding.org
The Plan Mode / Debug Mode / Review Mode differentiation is thoughtful — it acknowledges that different tasks require different levels of agent autonomy and different kinds of human oversight. Giving the developer explicit control over which mode is active is a meaningful design choice.
The Agile Vibe Coding Manifesto's principle that "humans remain accountable for software systems" is directly relevant here: accountability doesn't mean doing everything yourself, but it does mean maintaining enough understanding and control to be genuinely responsible for the outcomes. Different agent modes with different oversight requirements is one way to structure that.
The "rules to inject repository conventions" feature is particularly interesting for teams: https://agilevibecoding.org
We’re building BitFun, an open-source Agentic Development Environment (ADE).
GitHub: https://github.com/GCWing/BitFun
Docs: https://deepwiki.com/GCWing/BitFun
BitFun explores a different way of working with AI during development. Instead of treating AI as a chatbox attached to an editor, it treats AI as an agent that can collaborate with you across the whole development workflow.
The system is built around an Agent architecture. Each user has an assistant that can evolve over time with memory and customizable behavior. On top of this foundation, BitFun currently includes:
Code Agent – an AI coding assistant
Cowork Agent – an AI assistant for knowledge work
Custom Agents – domain-specific agents defined via Markdown
For software development, the Code Agent provides several working modes:
Agentic Mode – day-to-day coding where the agent can read code, edit files, run commands, and verify results autonomously.
Plan Mode – for large tasks. The agent proposes a plan first, then executes after alignment.
Debug Mode – for tricky bugs. It instruments execution, traces code paths, compares behaviors, and helps identify root causes.
Review Mode – performs code reviews based on repository conventions and project rules.
BitFun is also designed to be extensible:
MCP protocol support for connecting external tools and resources
Skills: Markdown/script-based capability packages that teach the agent new tasks
Agent customization via Markdown definitions
Rules to inject repository conventions or behavior constraints (can import configs from tools like Cursor)
For model infrastructure, BitFun supports both local models and cloud APIs, so teams can switch providers easily depending on cost, performance, or privacy needs.
By default, code is not sent to remote services automatically. Users control which models run and where, and local models can run with full privacy.
Right now the desktop app (built with Rust + TypeScript / Tauri) supports macOS and Windows. We’re also exploring CLI, server mode, and integrations with platforms like Telegram or Discord.
This project started as an open-source exploration of human–AI collaborative development, and a large portion of the codebase was built through AI-assisted coding (“vibe coding”).
We’d love feedback from the community. Issues, ideas, and contributions are all welcome.