design-ux-architect
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because its workflow involves reading untrusted content from the file system to determine technical architecture and UX structure. If the ingested project files contain malicious instructions, they could influence the agent's output or behavior.
- Ingestion points: The agent is instructed to read
ai/memory-bank/site-setup.mdand task lists inai/memory-bank/tasks/usingcatandgrepcommands. - Boundary markers: Absent. The skill does not define delimiters or specific instructions for the agent to ignore potentially malicious embedded content within the source files.
- Capability inventory: The agent has the capability to read local files and generate executable code (CSS, JavaScript, HTML) which may be used by other agents or developers.
- Sanitization: Absent. There is no validation or filtering of the content read from the external files.
- [COMMAND_EXECUTION]: The skill workflow utilizes standard shell commands (
cat,grep) to analyze the project environment. While used for context gathering, these commands provide the agent with direct read access to the local file system. - [SAFE]: The provided JavaScript and CSS templates are standard implementations for theme management and design systems with no evidence of malicious behavior or obfuscation.
Audit Metadata