technical-blog-writing

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md Quick Start explicitly shows running "infsh app run exa/search" to "Research topic depth" and repeatedly instructs citing/following external "Further reading" links and platform posts, which indicates the skill fetches and ingests open web/third‑party content that can influence writing decisions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly requires and calls the inference.sh CLI (https://inference.sh) at runtime (e.g., "infsh app run ...", including infsh/python-executor), which causes remote execution of code and generation of prompts on the inference.sh platform, making it a required external runtime dependency that executes remote code.