AI agents call get_recent_prompts to retrieve information from Learnlog 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 learning history data (prompts from recent days). It has no capability to modify, delete, or execute external operations. It is a straightforward read-only operation on stored user prompts, consistent with the server's stated purpose of reviewing study history.
From the tool's definition Tool name is 'get_recent_prompts' and description translates to 'Retrieves questions from recent days' — a query operation that returns historical data without modification or side effects.
Attacks that exploit this kind of access
최근 며칠간의 질문들을 조회합니다. It is categorised as a Read tool in the Learnlog MCP Server, which means it retrieves data without modifying state.
Register the Learnlog MCP server in PolicyLayer and add a rule for get_recent_prompts: 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 Learnlog. Nothing to install.
get_recent_prompts 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_recent_prompts 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_recent_prompts. 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_recent_prompts is provided by the Learnlog MCP server (yujaeyun/learnlog-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 →