AI agents use research_import to create or update resources in Notebooklm — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Notebooklm environment.
This tool modifies notebook state by adding external research sources, which is a reversible write operation. It does not execute code, delete data, or move money. The medium severity reflects that importing untrusted or malicious sources could influence downstream analysis, but the operation itself is reversible and the blast radius is contained to notebook content.
From the tool's definition Tool name 'research_import' and description 'Import discovered research sources into a notebook' indicates creation/addition of data (sources) into a notebook resource.
Documented attack patterns abuse exactly the kind of access research_import gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Notebooklm, and nothing reaches the server without passing your rules. This is the rule we recommend for research_import:
{
"version": "1",
"default": "deny",
"tools": {
"research_import": {
"limits": [
{
"counter": "research_import_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}research_import stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Import discovered research sources into a notebook. It is categorised as a Write tool in the Notebooklm MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Notebooklm MCP server in PolicyLayer and add a rule for research_import: 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 Notebooklm. Nothing to install.
research_import 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 research_import 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 research_import. 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.
research_import is provided by the Notebooklm MCP server (moodrobotics/notebooklm-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Notebooklm, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
29 Notebooklm tools catalogued and risk-classified — across an index of 43,000+ MCP servers.