AI agents call debug_token to retrieve information from Meta without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and returns metadata about an authentication token without modifying any data, executing code, or triggering side effects. It is a straightforward inspection/debugging operation. Low severity because token metadata exposure alone does not enable unauthorized access or destructive actions (the token itself would be needed for that), though it could inform reconnaissance.
From the tool's definition Tool name is 'debug_token' and description states it 'returns token type, app_id, expires_at, scopes, is_valid' — a pure retrieval operation with no mutation or execution capability. The HTTP method is GET. Server is explicitly described as 'Read-only'.
Attacks that exploit this kind of access
GET /debug_token — returns token type (USER / PAGE / APP / SYSTEM_USER), app_id, expires_at, scopes, is_valid. It is categorised as a Read tool in the Meta MCP Server, which means it retrieves data without modifying state.
Register the Meta MCP server in PolicyLayer and add a rule for debug_token: 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 Meta. Nothing to install.
debug_token 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 debug_token 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 debug_token. 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.
debug_token is provided by the Meta MCP server (nourpups/meta-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 →