Run a shell command inside the sandbox. Returns combined stdout+stderr. Use for any CLI work: git, npm, pip, apt, curl, etc.
AI agents invoke exec to trigger actions in Taw Computer. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool allows unrestricted execution of shell commands in a sandboxed Ubuntu environment. While sandboxing provides some isolation, shell command execution is inherently Execute category (not Read, Write, or Destructive alone) because its effects depend entirely on the command arguments.
From the tool's definition 'Run a shell command inside the sandbox' - this directly executes arbitrary shell commands with full system access, including dangerous operations like package installation (apt), file operations, git, npm, pip, curl, etc.
Documented attack patterns abuse exactly the kind of access exec gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Taw Computer, and nothing reaches the server without passing your rules. This is the rule we recommend for exec:
{
"version": "1",
"default": "deny",
"tools": {
"exec": {
"limits": [
{
"counter": "exec_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}exec stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Run a shell command inside the sandbox. Returns combined stdout+stderr. Use for any CLI work: git, npm, pip, apt, curl, etc. It is categorised as a Execute tool in the Taw Computer MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Taw Computer MCP server in PolicyLayer and add a rule for exec: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Taw Computer. Nothing to install.
exec is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the exec rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for exec. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
exec is provided by the Taw Computer MCP server (tawgroup/taw-computer). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Taw Computer, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
36 Taw Computer tools catalogued and risk-classified — across an index of 43,000+ MCP servers.