systems-thinking
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [Prompt Injection] (SAFE): No attempts to override system instructions or bypass safety filters were detected. The skill uses standard instructional language for organizational analysis.
- [Data Exposure & Exfiltration] (SAFE): The skill does not perform any network operations or access sensitive local file paths. It operates entirely on the data provided by the user within the chat context.
- [Unverifiable Dependencies] (SAFE): There are no external packages or scripts referenced or installed by this skill. It is a 'no-code' skill consisting only of Markdown and metadata.
- [Indirect Prompt Injection] (LOW): The skill is designed to process untrusted user input (e.g., problem statements, stakeholder lists). However, since the skill has no destructive capabilities like file-writing, code execution, or network access, the risk of an indirect prompt injection attack causing harm is negligible.
Audit Metadata