AI agents call cvat_get_api_schema to retrieve information from CVAT MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves schema documentation via a GET request, which is a passive, informational operation. There are no side effects, no data modification, no code execution, and no irreversible actions. The purpose is to inspect available API endpoints, making it a straightforward Read category tool.
From the tool's definition Tool description explicitly states 'Fetch the CVAT OpenAPI schema from GET /api/schema/' — a read-only retrieval operation that inspects API metadata without modifying or executing operations.
Attacks that exploit this kind of access
Fetch the CVAT OpenAPI schema from GET /api/schema/ so official API endpoints can be inspected or converted into MCP tools. It is categorised as a Read tool in the CVAT MCP MCP Server, which means it retrieves data without modifying state.
Register the CVAT MCP server in PolicyLayer and add a rule for cvat_get_api_schema: 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 CVAT MCP. Nothing to install.
cvat_get_api_schema 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 cvat_get_api_schema 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 cvat_get_api_schema. 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.
cvat_get_api_schema is provided by the CVAT MCP server (jangjs1216/cvat-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 →