$ loading_
Give AI agents authenticated sandbox access and a restricted path for safe production deploys.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "mcp-sandbox-server" yet — see the docs or source repo.
Connect to mcp-sandbox-server, inspect the /app directory on the sandbox host, find obvious issues in the configuration files, and suggest fixes.
A report of inspected files and directories, suspicious configuration items, and suggested fixes or next steps.
Use the restricted deploy path in mcp-sandbox-server to release the current version to the production host, and explain the deployment actions before running them.
An explanation of the allowed deployment steps first, followed by the deployment result or failure details.
Use mcp-sandbox-server to run basic diagnostic commands on the sandbox host, verify shell access works, and summarize the results.
The command execution status, key output, and a brief conclusion about current accessibility.
Developers or DevOps teams can let an AI agent access a sandbox host with authenticated shell and file capabilities to inspect code, configuration, or environment issues. This enables analysis in isolation before taking further action.
When AI needs to participate in releases, this tool provides a restricted gateway to the production host and limits actions through forced SSH commands. It fits deployment workflows that require stronger safety boundaries.
It gives an AI agent authenticated shell and file access to a sandbox host, plus a restricted gateway for safer deployment to a production host. The description specifically notes forced SSH commands on the production side.
Based on the description, production access is a restricted gateway that relies on forced SSH commands, so it is not unrestricted general access. For the exact allowed commands, see the source repository.
Known prerequisites include an accessible sandbox host and the setup needed for authenticated access and SSH-restricted deployment. For exact installation steps, runtime requirements, or key configuration, see the source repository.
Run commands, manage long jobs, and transfer files in AI sandboxes.
A sandbox server for testing and debugging MCP tools and interactions.
Securely run Python code in a sandbox and return execution results.
Run code and Postman collections in isolated sandboxes with real-time management.
Verify MCP server responses by returning a unique canary identity string.
Manage AWS sandbox accounts, leases, templates, events, and settings with natural language.