systematic-debugging

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The file find-polluter.sh contains a bash script that programmatically executes npm test on a series of files identified through user-defined search patterns.
  • [COMMAND_EXECUTION]: Documentation in SKILL.md and root-cause-tracing.md provides examples of shell commands for diagnostic instrumentation, such as inspecting environment variables and system keychain identities.
  • [PROMPT_INJECTION]: The skill uses highly imperative language and 'Iron Laws' (e.g., 'NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST') to override the agent's default behavior, mandating a specific multi-phase process for debugging tasks.
  • [COMMAND_EXECUTION]: The skill includes logic for condition-based waiting in TypeScript (condition-based-waiting-example.ts), which is an automated testing capability used to prevent race conditions during execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 03:11 PM