technical-blog-writing

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md Quick Start explicitly instructs using the CLI search app (infsh app run exa/search) to "Research topic depth", which implies fetching and ingesting open web/search results (untrusted third‑party content) that the agent is expected to read and that could materially 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 instructs installing and using the inference.sh CLI via "curl -fsSL https://cli.inference.sh | sh" and then running remote apps (e.g., infsh app run infsh/python-executor) which execute arbitrary code / control agent behavior, so https://cli.inference.sh is a runtime dependency that executes remote code.