BitFun is an open-source AI coding agent and developer execution platform maintained by GCWing and community contributors. Built on a modular Rust-backed runtime and multi-package architecture, it provides an agent harness capable of coordinating automated programming workflows across local environments and self-hosted server instances. In daily use, BitFun connects to multiple LLM providers via configurable routes and API credentials, supporting advanced models such as Claude and OpenAI-compatible engines. The platform leverages an internal app-server runtime to orchestrate profile-scoped agent catalogs, manage atomic tool groups, and execute tasks across terminal interfaces, headless server deployments, or embedded surfaces. BitFun differentiates itself by decoupling agent orchestration into atomic tool clusters and headless execution layers, featuring intelligent multi-route model deduplication and dedicated policy flows such as DeepResearch. This modular separation allows developers to deploy the runtime as a headless backend or connect it to web and embedded mini-app interfaces. The project is open source and distributed through GitHub under rapid version 0.2.x development, offering installer packages and release synchronization scripts for server and desktop deployments.
Multi-provider route orchestration — Maps and deduplicates model availability across distinct API endpoints and credentials within a unified catalog. - Headless runtime architecture — Runs agent sessions headlessly with scoped profiles and isolated tool boundaries suited for CI/CD and server environments. - Atomic tool grouping — Organizes execution capabilities into discrete tool groups to prevent context pollution and enforce execution boundaries. - Integrated app server — Provides a standalone backend service to manage persistent sessions, agent coordination, and client communication. - Embedded surface rendering — Supports projecting interface rendering across embedded surfaces and mini-app web components. - DeepResearch workflow policies — Configures specialized policy rules for complex multi-step analysis and research tasks.
Use cases
Use Case: Multi-Provider LLM Routing Developers managing multiple API subscriptions can configure separate provider routes for models such as Claude without duplicate listings in the session catalog. BitFun resolves the correct credentialed route while keeping the active model session consistent across interactions.
Use Case: Headless Agent Task Execution Platform teams can deploy BitFun on headless servers or CI runners using atomic tool groups and scoped runtime profiles. The agent executes automated code modifications, refactors, and diagnostic tasks without requiring an interactive desktop UI.
Use Case: Automated Research and Refactoring Engineers can initiate multi-step DeepResearch tasks to analyze repository architecture across packages before applying patches, ensuring workflow policies and dependencies are validated prior to execution.
Who it is for
BitFun is designed for systems developers, platform engineers, and AI infrastructure teams who need a self-hostable, multi-provider agent harness capable of headless or server-backed execution. It is less suitable for users looking for a zero-configuration, single-binary CLI tool with out-of-the-box cloud subscriptions.
Install guide
Installation Clone the repository and build using Cargo (standard Rust workflow based on the repository structure): bash git clone https://github.com/GCWing/BitFun.git cd BitFun cargo build --release Alternatively, use the included installer component: bash cd BitFun-Installer # Follow platform-specific build instructions
First Run Configure your provider API keys and runtime parameters in your environment or configuration file: bash export ANTHROPIC_API_KEY="your-anthropic-key" export OPENAI_API_KEY="your-openai-key" # Start the BitFun app server ./target/release/bitfun app-server
Example Session Run an agent task against your local workspace: bash bitfun run --profile default --task "Analyze the dsh-acp package and refactor model route deduplication logic"
Plandex — Plandex is an open-source terminal-based AI coding agent designed to handle complex, multi-file engineering tasks in large codebases. Created by…
OpenCode — OpenCode is an open-source AI coding agent developed by Anomaly that operates across terminal environments, desktop applications, and IDE extensions.…
OpenHands — OpenHands (formerly OpenDevin) is an open-source autonomous AI software development agent and orchestration platform. It provides an extensible runtime environment…
Subtask — Subtask is an open-source CLI agent harness and orchestration tool developed by zippoxer in Go. It is built to coordinate…
EchoCoding — EchoCoding is an open-source agent harness and audio-augmented developer tooling layer created by launsion-boop. Designed to run alongside AI coding…
pi-reflect — pi-reflect is an open-source agent harness and self-improvement extension for pi coding agents, developed by jo-inc. It inspects agent session…