recon-breweries

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document contains deliberate, actionable instructions to bypass protections and harvest sensitive data (API keys, customer/order data, coupons) from target sites, enabling data exfiltration and credential theft.

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md’s required runtime workflow uses curl to fetch arbitrary target web pages and JSON endpoints (public web content from the operating user’s chosen target), which is outsider-authored free text/structured prose that can be ingested into the agent’s LLM context if the tool output is passed through.