AI agents use update_ref to create or update resources in Substrate — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Substrate environment.
The tool updates (modifies) existing references, which is a Write operation. It is not Destructive because updates are typically reversible (the old value is replaced but not permanently lost in most reference systems). Severity is medium because unintended updates to references could break system functionality or configurations, but the impact is typically recoverable through subsequent updates or rollbacks.
From the tool's definition Tool name 'update_ref' and description 'Update existing reference' indicate modification of existing data in a reversible manner.
Attacks that exploit this kind of access
Update existing reference. It is categorised as a Write tool in the Substrate MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Substrate MCP server in PolicyLayer and add a rule for update_ref: 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 Substrate. Nothing to install.
update_ref 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_ref 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_ref. 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_ref is provided by the Substrate MCP server (ivan-saorin/substrate). 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 →