AI agents invoke install_and_restart to trigger actions in Unlimited. 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.
Installing packages modifies the system environment and introduces arbitrary third-party code, while restarting the server disrupts running services. Both actions have significant side effects beyond simple writes, qualifying as Execute with high severity due to the potential for supply-chain attacks and service disruption.
From the tool's definition 'Install a Python package with uv/pip then restart the server' — installs software and restarts a running service
Attacks that exploit this kind of access
Install a Python package with uv/pip then restart the server. It is categorised as a Execute tool in the Unlimited MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Unlimited MCP server in PolicyLayer and add a rule for install_and_restart: 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 Unlimited. Nothing to install.
install_and_restart 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 install_and_restart 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 install_and_restart. 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.
install_and_restart is provided by the Unlimited MCP server (triumsebas/unlimited-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →