agency-ux-architect
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill's workflow instructions direct the agent to execute shell commands such as
catandgrepto retrieve project context from the local filesystem (specifically within theai/memory-bank/directory). While these commands are scoped to specific paths, they represent an active interaction with the local environment. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface (Category 8). It ingests data from external project files which are then used to influence its reasoning and code generation outputs.
- Ingestion points: The agent reads from
ai/memory-bank/site-setup.mdand files withinai/memory-bank/tasks/using shell tools. - Boundary markers: There are no defined delimiters or instructions provided to the agent to help it distinguish between legitimate data and potential instructions embedded within the ingested files.
- Capability inventory: The agent possesses the capability to generate and suggest implementation-ready code, including CSS systems and JavaScript theme management logic, which are based on the ingested content.
- Sanitization: The skill does not provide any mechanisms for sanitizing, validating, or escaping the content retrieved from the local memory bank files before processing it.
Audit Metadata