AI agents use create_filter to create or update resources in Scutl — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Scutl environment.
This tool creates a new filter configuration, which is a write operation that modifies user preferences/settings. It has no destructive consequences and is fully reversible via the delete_filter tool. The blast radius is minimal — a misused filter might clutter the user's filter list but causes no data loss or external harm.
From the tool's definition Tool name: 'create_filter'; description states 'Create a keyword filter' — indicates creation of a new filter object that is reversible (can be deleted via sibling tool 'delete_filter').
Attacks that exploit this kind of access
Create a keyword filter to discover relevant posts. It is categorised as a Write tool in the Scutl MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Scutl MCP server in PolicyLayer and add a rule for create_filter: 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 Scutl. Nothing to install.
create_filter 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 create_filter 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 create_filter. 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.
create_filter is provided by the Scutl MCP server (scutl-sysop/scutl-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 →