message-test-designer
Pass
Audited by Gen Agent Trust Hub on Jul 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill proactively addresses indirect prompt injection risks. It includes specific instructions to treat user-provided stimulus sets, message variants, and panel notes as untrusted input, explicitly directing the agent to ignore any instructions embedded within that data.
- [SAFE]: Analysis of the skill's data handling shows it is restricted to local file operations within a managed
memory/directory structure. There are no patterns suggesting data exfiltration, network requests, or access to sensitive system credentials. - [SAFE]: The skill does not utilize any dynamic code execution, remote script downloads, or privilege escalation mechanisms. All logic is contained within the markdown instructions and focuses on generating a structured design specification.
Audit Metadata