AI agents invoke run_command to trigger actions in Search. 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.
Arbitrary command execution ('run_command') is inherently an Execute-category risk with critical severity because an AI agent could use it to execute malicious code, access files, modify system state, or perform unauthorized actions depending on the execution context and permissions.
From the tool's definition Tool name 'run_command' paired with description 'Run a command on this' indicates arbitrary command execution capability.
Documented attack patterns abuse exactly the kind of access run_command gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Search, and nothing reaches the server without passing your rules. This is the rule we recommend for run_command:
{
"version": "1",
"default": "deny",
"tools": {
"run_command": {
"limits": [
{
"counter": "run_command_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}run_command 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 command on this. It is categorised as a Execute tool in the Search MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Search MCP server in PolicyLayer and add a rule for run_command: 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 Search. Nothing to install.
run_command 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 run_command 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 run_command. 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.
run_command is provided by the Search MCP server (@agent-infra/mcp-server-search). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Search, 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 Search tools catalogued and risk-classified — across an index of 43,000+ MCP servers.