The open source AI coding agent - Enhanced by BreakingCircuits.
English | 简体中文 | 繁體中文 | 한국어 | Deutsch | Español | Français | Italiano | Dansk | 日本語 | Polski | Русский | Bosanski | العربية | Norsk | Português (Brasil) | ไทย | Türkçe | Українська
This is a BrokenCircuits Edition fork in early beta. Things may break.
This fork includes BC-specific customizations:
- Custom Skills: Memory management, Azure LLM bridge, Proxmox management, superpowers
- Enhanced Memory System: Short-term, medium-term, and long-term memory for session continuity
- Azure Integration: Built-in support for Azure AI Foundry (Mistral Large 3, Kimi K2)
- Proxmox Management: Native VM management for Dell PowerEdge R720 XD
- Security Focus: Tools for security research and penetration testing workflows
- Voice I/O: Text-to-speech and speech-to-text for hands-free operation
- Local AI: Ollama integration for free offline LLM inference
- Self-Improving: Magnitude framework that learns from every session
This fork includes a multi-file memory system for session continuity:
# Load all memory files at session start
python3 /home/bc/bin/memory.pyMemory hierarchy:
- CLAUDE.md (Long-term) - AI identity, user preferences, infrastructure
- MEDIUM_TERM.md (Cross-session) - Project progress, recent learnings
- SHORT_TERM.md (Current session) - Active tasks, conversation context
The memory loader prints all files in hierarchical order. Always run at session start!
| Skill | Description |
|---|---|
memory |
Remember/recall information across conversations with /remember, /recall, /forget |
azureridge |
CLI-llm-b interface to Mistral 3 Large and Kimi K2 via Azure AI Foundry |
azure-cli |
Azure resource management, credit monitoring, and deployment scripts |
proxmox-manager |
VM lifecycle management for Proxmox VE (list, create, status, delete VMs) |
| Skill | Description |
|---|---|
using-superpowers |
Skill invocation methodology - when to use skills |
brainstorming |
Creative work, feature design, and requirements exploration |
test-driven-development |
TDD workflow - tests before implementation |
systematic-debugging |
Bug investigation with root-cause tracing |
receiving-code-review |
Handling code review feedback professionally |
requesting-code-review |
Conducting effective code reviews |
using-git-worktrees |
Isolated workspace creation for feature development |
writing-plans |
Creating detailed implementation plans |
writing-skills |
Creating, testing, and verifying new skills |
subagent-driven-development |
Multi-agent execution for complex tasks |
executing-plans |
Plan execution with review checkpoints |
finishing-a-development-branch |
Branch completion, PR creation, cleanup |
verification-before-completion |
Pre-commit verification and testing |
dispatching-parallel-agents |
Parallel task execution for independent work |
| Skill | Description |
|---|---|
ace-context-plan |
ACE (Agentic Context Engineering) implementation plan for self-improving contexts |
magnitude |
Self-improvement framework with multi-agent research, vector memory, and autonomous learning (see plan) |
# YOLO
curl -fsSL https://opencode.ai/install | bash
# Package managers
npm i -g opencode-ai@latest # or bun/pnpm/yarn
scoop install opencode # Windows
choco install opencode # Windows
brew install anomalyco/tap/opencode # macOS and Linux (recommended, always up to date)
brew install opencode # macOS and Linux (official brew formula, updated less)
sudo pacman -S opencode # Arch Linux (Stable)
paru -S opencode-bin # Arch Linux (Latest from AUR)
mise use -g opencode # Any OS
nix run nixpkgs#opencode # or github:anomalyco/opencode for latest dev branchTip
Remove versions older than 0.1.x before installing.
OpenCode is also available as a desktop application. Download directly from the releases page or opencode.ai/download.
| Platform | Download |
|---|---|
| macOS (Apple Silicon) | opencode-desktop-darwin-aarch64.dmg |
| macOS (Intel) | opencode-desktop-darwin-x64.dmg |
| Windows | opencode-desktop-windows-x64.exe |
| Linux | .deb, .rpm, or AppImage |
# macOS (Homebrew)
brew install --cask opencode-desktop
# Windows (Scoop)
scoop bucket add extras; scoop install extras/opencode-desktopThe install script respects the following priority order for the installation path:
$OPENCODE_INSTALL_DIR- Custom installation directory$XDG_BIN_DIR- XDG Base Directory Specification compliant path$HOME/bin- Standard user binary directory (if it exists or can be created)$HOME/.opencode/bin- Default fallback
# Examples
OPENCODE_INSTALL_DIR=/usr/local/bin curl -fsSL https://opencode.ai/install | bash
XDG_BIN_DIR=$HOME/.local/bin curl -fsSL https://opencode.ai/install | bashTo build the BrokenCircuits Edition from source:
# Clone the fork
git clone git@github.com:breakingcircuits1337/opencodeBC-MAIN.git
cd opencodeBC-MAIN
# Install dependencies
bun install
# Build BC edition
cd packages/opencode
OPENCODE_CHANNEL=bc bun run buildThe binary will be at: dist/opencode-linux-x64/bin/opencode-bc
See BUILD.md for detailed build instructions and troubleshooting.
OpenCode includes two built-in agents you can switch between with the Tab key.
- build - Default, full-access agent for development work
- plan - Read-only agent for analysis and code exploration
- Denies file edits by default
- Asks permission before running bash commands
- Ideal for exploring unfamiliar codebases or planning changes
Also included is a general subagent for complex searches and multistep tasks.
This is used internally and can be invoked using @general in messages.
Learn more about agents.
For more info on how to configure OpenCode, head over to our docs.
If you're interested in contributing to OpenCode, please read our contributing docs before submitting a pull request.
If you are working on a project that's related to OpenCode and is using "opencode" as part of its name, for example "opencode-dashboard" or "opencode-mobile", please add a note to your README to clarify that it is not built by the OpenCode team and is not affiliated with us in any way.
It's very similar to Claude Code in terms of capability. Here are the key differences:
- 100% open source
- Not coupled to any provider. Although we recommend the models we provide through OpenCode Zen, OpenCode can be used with Claude, OpenAI, Google, or even local models. As models evolve, the gaps between them will close and pricing will drop, so being provider-agnostic is important.
- Out-of-the-box LSP support
- A focus on TUI. OpenCode is built by neovim users and the creators of terminal.shop; we are going to push the limits of what's possible in the terminal.
- A client/server architecture. This, for example, can allow OpenCode to run on your computer while you drive it remotely from a mobile app, meaning that the TUI frontend is just one of the possible clients.