agency-ux-architect
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading project-specific files which creates a surface for indirect prompt injection if those files contain malicious instructions.
- Ingestion points: The skill uses shell commands like
catandgrepto read content fromai/memory-bank/site-setup.mdand various files withinai/memory-bank/tasks/. - Boundary markers: There are no instructions or delimiters provided to ensure the agent ignores or treats instructions found within these files as data rather than commands.
- Capability inventory: The agent is capable of generating CSS, HTML, and JavaScript, and defining system architectures and API contracts based on the data it reads.
- Sanitization: The skill does not specify any sanitization, filtering, or validation steps for the content ingested from the memory bank files.
Audit Metadata