AI agents use set_pattern_name to create or update resources in Flai — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Flai environment.
Renaming a pattern creates or modifies metadata but is fully reversible—the user can simply rename it again. This is a Write operation, not Destructive, as no data is lost or overwritten irreversibly. The blast radius is minimal: the worst outcome is a confusing pattern name that can be corrected. In the context of a music production tool, this poses low risk to the integrity of a project.
From the tool's definition Tool description states 'Rename a pattern', which is a modification operation. The FL Studio context shows this operates on music production data (patterns are fundamental FL Studio objects used to organize musical sequences).
Attacks that exploit this kind of access
Rename a pattern. It is categorised as a Write tool in the Flai MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Flai MCP server in PolicyLayer and add a rule for set_pattern_name: 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 Flai. Nothing to install.
set_pattern_name 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 set_pattern_name 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 set_pattern_name. 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.
set_pattern_name is provided by the Flai MCP server (kaupau/flai-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 →