AI agents invoke RunPython to trigger actions in OfficeMCP. 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.
RunPython executes code whose effects depend entirely on the arguments provided. An AI agent could use this to run malicious scripts, exfiltrate data, modify files, or compromise the system. The empty description increases risk since capabilities are not explicitly constrained.
From the tool's definition Tool name 'RunPython' indicates execution of arbitrary Python code. In the context of an OfficeMCP server that automates Microsoft Office via COM interface, this tool can execute arbitrary Python scripts that interact with Office applications and the…
Documented attack patterns abuse exactly the kind of access RunPython gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and OfficeMCP, and nothing reaches the server without passing your rules. This is the rule we recommend for RunPython:
{
"version": "1",
"default": "deny",
"tools": {
"RunPython": {
"limits": [
{
"counter": "runpython_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}RunPython 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.
RunPython. It is categorised as a Execute tool in the OfficeMCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Office MCP server in PolicyLayer and add a rule for RunPython: 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 OfficeMCP. Nothing to install.
RunPython 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 RunPython 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 RunPython. 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.
RunPython is provided by the Office MCP server (officemcp/officemcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 16 OfficeMCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
16 OfficeMCP tools catalogued and risk-classified — across an index of 42,500+ MCP servers.