OpenCode Most — AI skill for Claude Code
Use any OpenCode Zen model (Muse Spark, Big Pickle, Ling, MiMo, Nemotron.
How to install OpenCode Most
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open tavrida1337-sketch/OpenCode-Most and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What OpenCode Most does
Use any OpenCode Zen model (Muse Spark, Big Pickle, Ling, MiMo, Nemotron...) from Claude Code or any OpenAI client - free, local, no API keys
Alternatives in AI
- Zen MCP Server — The power of Claude Code + Gemini / OpenAI / Grok / OpenRouter / Ollama / Custom Model working as one 11.3k ★
- Vibeproxy — Native macOS menu bar app to use your Claude Code & ChatGPT subscriptions with AI coding tools - no API keys n 3.3k ★
- EchoBird — One-click install + model switch:Claude Code,Codex CLI (OpenAI), Grok Build (xAI), DeepSeek Harness, Kimi Code 3.1k ★
README
OpenCode Bridge 🌉
**Use any OpenCode Zen model (Muse Spark, Big Pickle, Ling, MiMo, Nemotron…) from Claude Code or any OpenAI client — free, local, no API keys.**
[Русская версия](https://github.com/tavrida1337-sketch/OpenCode-Most#%D1%80%D1%83%D1%81%D1%81%D0%BA%D0%B8%D0%B9-)
What is this?
OpenCode Desktop runs a hidden local server with all your chats and models. **OpenCode Bridge** finds that server automatically and exposes it as:
- ✅ OpenAI-compatible API →
POST /v1/chat/completions(streaming supported) - ✅ Anthropic-compatible API →
POST /v1/messages(this is what Claude Code speaks) - ✅ Web dashboard → pick the session, pick the Zen model, test prompts, live status
The model that answers is **the one selected in OpenCode** — switch it in the app (or on the dashboard) and the next request follows instantly.
No dependencies. One folder. Pure Python stdlib.
✨ Features
| 🤖 Claude Code integration | Point it at the bridge, pick the OpenCode model, code with free Zen models |
| 🧠 Follows your picker | Change the model at the bottom of OpenCode → next request uses it; or force one from the site dropdown |
| 🌐 Zen model list | Big Pickle, Ling 3.0, MiMo V2.5, Muse Spark 1.2/1.3, Nemotron 3 Ultra/3.5… — same catalogue as the desktop picker |
| 📂 Session picker | Requests go into the session whose folder you have open — edits land in the right project |
| 🔄 Auto-discovery | Finds the server port (TCP table) + password (process env) by itself, re-discovers on rotation |
| 📊 Health page | Plain-text GET /status → status api:work / status api:no work (error: …) |
| 🛡️ Resilient | Retries fast network drops, polls for late answers, never crashes on upstream errors |
🚀 Quick start
- OpenCode Desktop must be running (any chat open).
- Run
start.bat— the bridge starts in the background, the dashboard opens athttp://127.0.0.1:8000/. - For Claude Code:
Related Skills
Agent Dispatch
Free skill for Claude Code, Cursor and any agent client: picks the model for each subagent by the type of task
Ocincc
Use OpenCode Zen's free AI models inside Claude Code — one command setup (npx ocincc@latest)
OpenFable
Terminal AI coding assistant with Mythos reasoning, Claude Code tools, free MiMo Auto models, OBLITERATUS abli
Opencode Starter
Interactive launcher for AI coding tools via OpenCode backends (Zen/Go)
Agent Network
助力搭建你的数字 AI 员工军团 — 多 Agent 一行命令组网协作。Claude Code / Claude Agent SDK / Codex / Grok Build 4 runtime + 8+ 家 LLM(A
Unblind
A vision backend for AI Agents — self-healing Claude Code Agent Skill. Routes images to Mimo/OpenAI/Ollama vis
Related Agents
Claude Advisor
Second-opinion advisor running Claude's most capable model (Fable 5). Consult at commitment boundaries — befor
LLM Orchestrator
Use this agent for LLM integration work — prompt engineering, multi-provider abstraction (OpenRouter, Gemini,
Provider Tester
Use this agent when you need to test, debug, or validate LLM provider configurations. This includes verifying