OpenCode
Open source coding agent for autonomous software development.
4 members list OpenCode, up 4 in the last 12 weeks, data as of 13 November 2026
Used alongside OpenCode
Tools that members list in the same stack as OpenCode.
Members who list OpenCode in their Stack, by week.
Members who list OpenCode
Homebrew analytics installs: 36,838 installs (2026) to 44,067 installs (2026)
Imported from Homebrew analytics, last reading Oct 2026
About OpenCode
OpenCode is an open source coding agent that helps developers write and run code in various environments including the terminal, desktop applications, and IDE extensions. The tool works with any AI model, offering free built-in models as well as integration with over 75 LLM providers such as Claude, GPT, Gemini, and local models. It supports features like multi-session parallel agents on the same project and integration with existing GitHub Copilot and ChatGPT Plus or Pro subscriptions.
Developed by anomalyco, OpenCode is released under the MIT License and is entirely free to use. The tool is written in TypeScript and runs on Windows, Unix-like systems, and macOS. It prioritizes privacy by design, not storing code or context data when operating in privacy-sensitive environments. The source code is publicly available on GitHub, allowing community members to file issues, submit pull requests, and contribute to its development.
Checked 8 November 2026.
Posts about OpenCode
- Claude Code alternatives: add Codex first, and the three agents that replace it
Three agents replace Claude Code as of November 2026, but 5 of 8 Stackness members who use it run a second agent beside it. Add Codex before you leave.
- Cursor Agent SDK vs Claude Code agents: what ports to another harness, and what stays on Cursor
Cursor's Agent SDK went GA on 20 October 2026. Its prompts, skills and MCP configs port; its triggers, hosting and model calls stay on Cursor.
- Claude Code status line token usage: the status line is free, agent_summary and startup context are not
Claude Code's status line uses 0 tokens. agent_summary took 2% to 24% of spend where measured. My startup context was 43,394 tokens.