svelte5-runes
Fail
Audited by Snyk on Feb 15, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (high risk: 0.80). The prompt includes hidden LLM workflow and editing instructions inside an HTML comment (e.g., "LLM WORKFLOW" and claude-skills-cli steps) that are unrelated to Svelte guidance and instruct the agent to change its behavior, which is a concealed prompt injection.
Audit Metadata