AI agents invoke reload_mcp_server to trigger actions in MCP Proxy. 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 likely triggers a reload/restart of an MCP server process, which is an external operation affecting running services. This falls under Execute as it triggers an external operational action. The description is empty, reducing confidence, but the server context of 'hot-reloading' MCP servers strongly implies this tool restarts or reinitializes a server process.
From the tool's definition Tool name 'reload_mcp_server' combined with server description mentioning 'hot-reloading' and 'orchestration of MCP servers'
Attacks that exploit this kind of access
reload_mcp_server. It is categorised as a Execute tool in the MCP Proxy MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MCP Proxy MCP server in PolicyLayer and add a rule for reload_mcp_server: 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 MCP Proxy. Nothing to install.
reload_mcp_server 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 reload_mcp_server 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 reload_mcp_server. 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.
reload_mcp_server is provided by the MCP Proxy MCP server (lizthedeveloper/mcp_proxy). 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 →