AI agents use create_paragraph to create or update resources in Zeppelin — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Zeppelin environment.
This tool creates new content (paragraphs) within a Zeppelin notebook, which is a reversible modification. It does not execute code, delete data, or move money. While it can be part of a workflow that leads to code execution via run_paragraph, the tool itself merely creates/adds content.
From the tool's definition Tool description states 'Add a new paragraph to a notebook', which is a create operation that modifies notebook content reversibly.
Attacks that exploit this kind of access
Add a new paragraph to a notebook. Optionally specify text and position index. It is categorised as a Write tool in the Zeppelin MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Zeppelin MCP server in PolicyLayer and add a rule for create_paragraph: 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 Zeppelin. Nothing to install.
create_paragraph 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_paragraph 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_paragraph. 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_paragraph is provided by the Zeppelin MCP server (mihneamanolache/zeppelin-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 →