AI agents call get_students to retrieve information from Datagsm without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves student information from a database with no capability to create, modify, delete, or execute actions. It is a straightforward read operation. Severity is low because querying student directory data, while potentially containing PII, does not grant destructive or financial capabilities and depends on proper backend access controls.
From the tool's definition Tool description states it queries a student list ('학생 목록을 조회합니다') with filtering options by grade, class, and major. The verb 조회 (to query/retrieve) and the context of DataGSM (an educational management system) confirm retrieval with no modification.
Attacks that exploit this kind of access
DataGSM에서 학생 목록을 조회합니다. 학년, 반, 전공 등 다양한 조건으로 필터링할 수 있습니다. It is categorised as a Read tool in the Datagsm MCP Server, which means it retrieves data without modifying state.
Register the Datagsm MCP server in PolicyLayer and add a rule for get_students: 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 Datagsm. Nothing to install.
get_students 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_students 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_students. 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_students is provided by the Datagsm MCP server (snowykte0426/datagsm-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 →