AI agents call suggest_metadata to retrieve information from Lessons without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves and displays existing metadata (categories and tags) to inform the user before they write a lesson. It has no side effects, creates no new data, and performs only a read/query operation. This is consistent with the 'Read' category for operations that retrieve or query data without side effects.
From the tool's definition Tool name is 'suggest_metadata' and description states it 'Preview available categories and tags before writing a new lesson' — this is a lookup/preview operation with no modification of data.
Attacks that exploit this kind of access
Preview available categories and tags before writing a new lesson. It is categorised as a Read tool in the Lessons MCP Server, which means it retrieves data without modifying state.
Register the Lessons MCP server in PolicyLayer and add a rule for suggest_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 Lessons. Nothing to install.
suggest_metadata 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 suggest_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 suggest_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.
suggest_metadata is provided by the Lessons MCP server (sosoyososo/lesson-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 →