skills/google/adk-python/adk-debug/Gen Agent Trust Hub

adk-debug

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • Local Service Interaction: The skill instructions involve using curl to interact with a local development server at http://localhost:8000. This is used to query session data, health status, and debug traces during the development process.
  • Piped Command Execution: Several examples demonstrate piping curl output to python3 -m json.tool. This pattern is used for formatting JSON data for readability. While piping to an interpreter is a pattern that warrants attention, the use of a standard formatting module on local data is a routine practice for developers.
  • Dynamic Script Generation: The skill suggests that the agent should generate and execute Python scripts to process and summarize JSON data retrieved from the debug server. This dynamic execution is intended to help troubleshoot agent behavior based on real-time data structures.
  • Local File Access: The skill describes monitoring log files located in /tmp/agents_log/ using the tail command. This provides transparency into the agent's internal operations and error states.
  • Configuration via Environment Variables: It references the use of environment variables such as GOOGLE_CLOUD_PROJECT and ADK_CAPTURE_MESSAGE_CONTENT_IN_SPANS to manage tracing and logging behavior, which are standard configuration methods.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 06:05 PM
Security Audit — agent-trust-hub — adk-debug