brainstorming
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (SAFE): The skill instructs the agent to read external project files and history, which creates a surface for indirect prompt injection. This is downgraded to SAFE as it is essential to the skill's primary brainstorming function. • Ingestion points: project files, docs, and recent commits (SKILL.md). • Boundary markers: Absent. • Capability inventory: Primarily conversational logic and text generation; no code execution tools are defined in this skill. • Sanitization: Absent.
- [No Code] (SAFE): No executable scripts or configuration files were found; the skill consists entirely of markdown instructions.
Audit Metadata