blog

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The prompt includes an explicit "Community Footer" instruction to append promotional content to conversation outputs after major deliverables, which is unrelated to the core blog-authoring functionality and constitutes a deceptive behavioral instruction outside the skill's stated purpose.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill explicitly performs WebSearch and fetches public URLs (see the blog-researcher agent: "Uses WebSearch to find current statistics, competitor content, and SERP analysis") and accepts/consumes external URLs via commands like /blog analyze <file-or-url> and research packets are then read and acted on by downstream agents (blog-writer, blog-seo), so untrusted third‑party content can materially influence tool decisions and next actions.