well-architected-framework-assessment
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, security risks, or safety violations were detected. The skill is an administrative tool that uses authorized Snowflake metadata queries to perform assessment tasks.
- [COMMAND_EXECUTION]: The skill utilizes the
bashtool for standard local environment management, such as creating output directories for reports and opening the generated HTML file in the system browser. These actions are user-initiated and restricted to the skill's stated purpose. - [DATA_EXFILTRATION]: All data gathered during the assessment is processed locally and stored in a report on the user's machine. No unauthorized external network requests or data transfers to remote servers were identified.
- [PROMPT_INJECTION]: The skill's instructions and subagent templates are structured for objective data analysis and grading. No attempts to bypass safety filters, ignore system instructions, or extract sensitive model information were found.
- [REMOTE_CODE_EXECUTION]: The skill relies on local YAML configuration and standard Snowflake SQL execution tools. It does not download external scripts or execute untrusted code from remote sources.
Audit Metadata