brainstorm
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill exhibits no malicious behavior and its tools are used appropriately for its intended research and ideation functions. It primarily focuses on generating creative possibilities and writing them to a local folder.
- [PROMPT_INJECTION]: The skill ingests untrusted data from the web which provides context for multiple agents, creating an indirect prompt injection surface. Evidence: 1. Ingestion points: Signal Scout fetches web content (SKILL.md). 2. Boundary markers: External data is interpolated directly into ideation agent prompts without delimiters. 3. Capability inventory: Agent spawning and file writing capabilities. 4. Sanitization: No sanitization of web content is specified in the prompt instructions.
Audit Metadata