AI agents invoke run_prompt to trigger actions in Security Copilot MCP Server. 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.
The tool executes prompts in an external security system (Microsoft Security Copilot). Prompt execution in AI/security platforms typically triggers workflows, data retrieval, or actions whose effects depend on the prompt content and the backend system's capabilities. This is Execute rather than Read because prompts can drive side effects (searches, alerting, remediation triggers) beyond simple data retrieval.
From the tool's definition Tool named 'run_prompt' with empty description, deployed on a Security Copilot MCP server that 'execute[s] prompts in Security Copilot.' Sibling tools include 'run_sentinel_query' (query execution) and 'upload_plugin' (code deployment), establishing this…
Documented attack patterns abuse exactly the kind of access run_prompt gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Security Copilot MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for run_prompt:
{
"version": "1",
"default": "deny",
"tools": {
"run_prompt": {
"limits": [
{
"counter": "run_prompt_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}run_prompt 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_prompt. It is categorised as a Execute tool in the Security Copilot MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Security Copilot MCP Server MCP server in PolicyLayer and add a rule for run_prompt: 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 Security Copilot MCP Server. Nothing to install.
run_prompt 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_prompt 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_prompt. 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_prompt is provided by the Security Copilot MCP Server MCP server (jguimera/securitycopilotmcpserver). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Security Copilot MCP Server, 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.
4 Security Copilot MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.