AI agents use create_request to create or update resources in Paraph — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Paraph environment.
This tool creates new signing requests and modifies documents by populating template data. While the action is reversible (requests can be cancelled via the sibling 'cancel_request' tool), it commits data into the system and initiates external workflows that bind signers to documents.
From the tool's definition Tool 'create_request' fills PDF templates and sends them for e-signing. The description explicitly states it can 'Fill a PDF template and optionally send it for e-signing,' which creates new document instances and initiates signing workflows.
Attacks that exploit this kind of access
Fill a PDF template and optionally send it for e-signing. Omit signers for fill-only. It is categorised as a Write tool in the Paraph MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Paraph MCP server in PolicyLayer and add a rule for create_request: 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 Paraph. Nothing to install.
create_request 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_request 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_request. 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_request is provided by the Paraph MCP server (servants-of-the-server-fire/paraph-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 →