zhongkao-xiaosimen

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The prompt includes a mandatory "publish to GitHub Pages" workflow (read files, run git add/commit/push, append README, expose a public URL) that forces network/IO actions outside the skill's stated purpose of generating practice questions, so it contains deceptive/out-of-scope instructions.