Update an installed MCP server to the latest version
AI agents use update_mcp to create or update resources in MCP Secure Installer — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MCP Secure Installer environment.
This tool modifies existing data/state (the installed MCP server) in a reversible manner. While updates could theoretically introduce breaking changes, the operation itself is a standard Write action (updating/replacing configuration or software state).
From the tool's definition Tool name 'update_mcp' and description 'Update an installed MCP server to the latest version' indicate modification of existing installed software. The action modifies the state of an installed MCP server by replacing it with a newer version.
Attacks that exploit this kind of access
Update an installed MCP server to the latest version. It is categorised as a Write tool in the MCP Secure Installer MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the MCP Secure Installer MCP server in PolicyLayer and add a rule for update_mcp: 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 Secure Installer. Nothing to install.
update_mcp is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the update_mcp 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 update_mcp. 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.
update_mcp is provided by the MCP Secure Installer MCP server (mossaka/mcp-sinstaller). 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 →