AI agents use update_metadata to create or update resources in Scrivener — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Scrivener environment.
This tool modifies document metadata reversibly (title, synopsis, label, status, includeInCompile). Changes can be undone by updating again with different values. It does not delete data (Destructive), execute code (Execute), move money (Financial), or merely read (Read).
From the tool's definition Tool description states it 'Updates title, synopsis, label, status, or includeInCompile for a document' — clear metadata modification without deletion or code execution.
Attacks that exploit this kind of access
Updates title, synopsis, label, status, or includeInCompile for a document. Will fail with a clear error if Scrivener has the project open — close Scrivener first. It is categorised as a Write tool in the Scrivener MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Scrivener MCP server in PolicyLayer and add a rule for update_metadata: 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 Scrivener. Nothing to install.
update_metadata 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_metadata 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_metadata. 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_metadata is provided by the Scrivener MCP server (sschmitt-cg/scrivener-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 →