opencode-sandbox
Spins up isolated opencode instances in Docker containers as MCP servers, providing 80 tools for file and shell operations within a scoped workspace without exposing the host filesystem.
README
opencode-sandbox
Spin up an isolated opencode instance from prebuilt public GHCR images and hand
it to Claude Code as an MCP server. No build step, no checkout of the opencode
fork — each instance is its own Docker Compose project with its own named
volumes, and its workspace is a named volume rather than a host bind, so it
cannot see your filesystem. Images are public and pull anonymously; the
backend reports opencode 1.18.4 and the MCP bridge exposes 80 tools.
Install
Build from source and drop the binary on your PATH:
bun run build
cp dist/opencode-sandbox ~/.local/bin/
(bun run install-local does both in one step.) Or grab a prebuilt binary
from this repo's GitHub Releases.
Quickstart
Create an instance. Any provider API key already in your shell
(ANTHROPIC_API_KEY, OPENAI_API_KEY, GOOGLE_GENERATIVE_AI_API_KEY,
OPENROUTER_API_KEY) is forwarded into it at create time — there's no UI
login step and no auth.json to manage.
opencode-sandbox create scratch --seed ./my-project --config ./opencode.jsonc
--seed copies a directory into the fresh workspace before the instance first
starts; --config places an opencode.jsonc for that instance alone. Both are
optional.
Register it with Claude Code. create prints this line, and url reprints it
later:
opencode-sandbox url scratch
claude mcp add --transport http scratch http://127.0.0.1:4784/mcp
Now use it — the instance's 80 tools (read, write, run shell commands, all
scoped to its own /workspace) are available in that Claude Code session.
When you're done:
opencode-sandbox destroy scratch --yes
destroy removes the containers and their volumes, including the workspace.
A web UI is available too, behind --web on create; without it, the MCP
bridge is the only interface.
Security note
The MCP bridge has no inbound authentication of its own. Anyone who can
reach its port gets the full opencode tool surface — reading and writing
files and running shell commands inside that instance's /workspace. The
published port is bound to 127.0.0.1 by default; keep it that way. Do not
widen it to 0.0.0.0 or put it behind a public reverse proxy without a real
auth layer in front of it.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.