gstack-investigate

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a guide for developers/agents to debug codebases. It outlines a logical workflow involving reproduction, hypothesis testing, and regression verification.
  • [DATA_EXFILTRATION]: The skill includes security-positive instructions that explicitly forbid the agent from including sensitive information (secrets, hostnames, IPs, customer data) in web searches when looking up error messages.
  • [COMMAND_EXECUTION]: Shell usage is limited to standard git operations (git status, git log) and the project's own testing tools, which is consistent with the skill's purpose of codebase investigation.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes external project data and test outputs, it does so within a debugging context. The instructions emphasize human review and architectural questioning if initial fixes fail, providing a manual oversight layer.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 09:43 AM