brainstorm
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No obfuscation, malicious code, or unauthorized network operations were detected. The skill's behavior aligns with its stated purpose as a brainstorming and design tool.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data through user-provided arguments and local file content. This is a standard risk for interactive agents and is categorized as safe given the lack of high-privilege capabilities. * Ingestion points: User input via the $ARGUMENTS variable and project data retrieved via Read, Glob, and Grep tools (SKILL.md). * Boundary markers: Absent; there are no specific delimiters to separate untrusted data from the agent's instructions. * Capability inventory: Access to Read, Glob, Grep, Task, and AskUserQuestion (SKILL.md). * Sanitization: No content filtering or validation is performed on the ingested data.
Audit Metadata