systematic-debugging
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): The skill uses imperative language ('The Iron Law', 'MUST complete each phase') to ensure the agent follows a disciplined debugging process. These instructions are functional constraints for the agent's task-specific behavior and do not involve bypassing safety guardrails or extracting system prompts.
- Indirect Prompt Injection (SAFE): While the skill is designed to ingest and analyze untrusted data (error logs and stack traces), the primary purpose of the skill is to facilitate investigation. The systematic approach (tracing data flow, verifying at multiple layers) acts as an inherent defense against accepting malicious instructions embedded in data.
- Command Execution (SAFE): The skill includes a shell script (
find-polluter.sh) and provides examples of diagnostic commands (e.g.,git init,codesign). These are standard development operations intended to be run in a local environment for debugging purposes and do not exhibit malicious intent.
Audit Metadata