The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the agent to execute a local Node.js or Python server (e.g., node .debug/debug_server.js) to capture runtime data.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection through the processed log files.
Ingestion points: The agent reads captured trace data from .debug/debug.log (SKILL.md, Step 7).
Boundary markers: The instructions do not specify the use of delimiters or protective warnings when the agent parses the log content.
Capability inventory: The agent has the ability to execute shell commands, modify project source code, and perform network requests.
Sanitization: There is no evidence of sanitization or filtering of the log data before it is processed by the agent.
[DATA_EXFILTRATION]: The provided debug servers in scripts/debug_server.js and scripts/debug_server.py are configured with Access-Control-Allow-Origin: *. This configuration allows any website visited in the user's browser to send POST requests to the local debug server while it is active, potentially injecting data into the logs.

interactive-debugging