design-ux-architect
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading and analyzing project requirements from local memory bank files (e.g., ai/memory-bank/site-setup.md and ai/memory-bank/tasks/*-tasklist.md). If these files contain data provided by untrusted sources, they could host malicious instructions that influence the agent's generated architecture or code.\n
- Ingestion points: Reads files in ai/memory-bank/ using cat and grep as specified in SKILL.md under Step 1 of the workflow.\n
- Boundary markers: The instructions do not define delimiters or specific 'ignore' rules for data read from these files.\n
- Capability inventory: The agent is authorized to use cat and grep for file reads and generates code deliverables like CSS and JavaScript foundations.\n
- Sanitization: There is no mention of sanitizing or validating the input data before it is used to generate foundation templates.\n- [COMMAND_EXECUTION]: The skill methodology explicitly directs the agent to use shell commands (cat, grep) to process information within its local environment. While intended for benign project analysis, it confirms the agent's operational mode of executing system-level commands to perform its duties.
Audit Metadata