Get detailed debugging information about the last error.
AI agents call jupyter_debug_last_error to retrieve information from ML Jupyter MCP 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 debugging information about a previously-occurred error. It performs no code execution, data modification, deletion, or external operations—it simply retrieves and displays diagnostic data about an existing error state.
From the tool's definition Tool name 'jupyter_debug_last_error' and description 'Get detailed debugging information about the last error' indicate a retrieval operation that provides diagnostic information without modifying state or executing code.
Attacks that exploit this kind of access
Get detailed debugging information about the last error. It is categorised as a Read tool in the ML Jupyter MCP MCP Server, which means it retrieves data without modifying state.
Register the ML Jupyter MCP server in PolicyLayer and add a rule for jupyter_debug_last_error: 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 ML Jupyter MCP. Nothing to install.
jupyter_debug_last_error 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 jupyter_debug_last_error 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 jupyter_debug_last_error. 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.
jupyter_debug_last_error is provided by the ML Jupyter MCP server (mayank-ketkar-sf/claudejupy). 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 →