brainstorming
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses imperative language in its description ('IMMEDIATELY USE THIS SKILL before answering any task') to force the AI agent into a specific operational mode for every interaction, attempting to override the agent's standard response protocol.
- [PROMPT_INJECTION]: Phase 2 instructions ('List all skills currently available in .claude/skills/') direct the agent to map its execution environment and available capabilities, which is a common reconnaissance step that can be used to identify further attack vectors.
Audit Metadata