bkend-cookbook

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to "use WebFetch" to load live documentation from public raw.githubusercontent.com URLs (e.g., the cookbooks and troubleshooting raw links), meaning the agent will fetch and interpret user-editable, third-party content that could influence its actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly directs using WebFetch to load raw markdown from https://raw.githubusercontent.com/popup-studio-ai/bkend-docs/main/en/cookbooks/blog/01-quick-start.md (and the sibling raw.githubusercontent.com URLs), which indicates these external files are fetched at runtime and could be injected into the agent context to control prompts/instructions.