$ loading_
Securely run shell commands for AI agents with process and output control.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Cuba-Exec" yet — see the docs or source repo.
Use Cuba-Exec to run "ls -la && df -h && uname -a" under a least-privilege policy, return only a concise summary of key results, and flag any blocked risky operations.
A summary of files, disk usage, and system info, plus whether all commands complied with policy.
Use Cuba-Exec to start a long-running script, record the process ID, capture bounded output continuously, and if it does not finish within 30 seconds, send SIGTERM first and SIGKILL if needed.
The result should include launch status, PID, truncated logs, and whether the process completed or was terminated.
Use Cuba-Exec to run "npm test" in the project directory, limit output length, extract key errors on failure, and avoid exposing irrelevant logs.
A concise result indicating pass/fail status, key error summaries, and minimal failure context.
Run safe, structured, stateful shell sessions on real infrastructure.
Run a single remote shell command safely and get summarized results.
Enable AI agents to run PowerShell, analyze scripts, and provide IntelliSense.
Lets AI agents run Python, execute scripts, and install pip packages locally.
Execute Lua, inspect scripts, and test Roblox client remotes.
Execute shell commands securely over MCP for controlled dev and ops workflows.