AI agents call debugpy_variables to retrieve information from Debugpy without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs inspection/querying of variable values during debug sessions. While it is Read category (no side effects), severity is medium rather than low because variable inspection during debugging can expose sensitive information (credentials, tokens, personal data) that might be present in memory, creating a potential information disclosure risk if an AI agent misuses it to extract secrets from a running…
From the tool's definition Tool name 'debugpy_variables' and description 'Inspect variables in a stopped frame' indicate a read-only operation that retrieves variable state from a debugging session without modifying code, data, or program state.
Attacks that exploit this kind of access
Inspect variables in a stopped frame. It is categorised as a Read tool in the Debugpy MCP Server, which means it retrieves data without modifying state.
Register the Debugpy MCP server in PolicyLayer and add a rule for debugpy_variables: 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 Debugpy. Nothing to install.
debugpy_variables 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 debugpy_variables 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 debugpy_variables. 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.
debugpy_variables is provided by the Debugpy MCP server (will-garrett/debugpy-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 →