get_notes
AI agents call get_notes to retrieve information from MCP AI Research Assistant without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries stored notes with no apparent side effects. It falls squarely within the Read category. Severity is low because retrieving notes poses minimal risk—no data is modified, deleted, or executed. The confidence is high (0.9) despite the empty description, as the tool name itself is sufficiently informative and the context of a research assistant supports this classification.
From the tool's definition Tool name is 'get_notes', which indicates retrieval of previously stored notes. The sibling tools include 'save_note' and 'summarize_text', consistent with a note-taking research assistant.
Attacks that exploit this kind of access
get_notes. It is categorised as a Read tool in the MCP AI Research Assistant MCP Server, which means it retrieves data without modifying state.
Register the MCP AI Research Assistant MCP server in PolicyLayer and add a rule for get_notes: 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 MCP AI Research Assistant. Nothing to install.
get_notes is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_notes 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 get_notes. 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.
get_notes is provided by the MCP AI Research Assistant MCP server (paneri11/mcp-ai-research-assistant). 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 →