press-release-writing

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

• Third-party content exposure detected (high risk: 1.00). SKILL.md's "Research & Fact-Checking" section explicitly instructs running inference.sh search apps (e.g., "infsh app run tavily/search-assistant" and "infsh app run exa/search") to fetch and interpret open web search results for verification and context, which are untrusted third-party sources that the agent will read and that can materially influence its outputs.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs users to run curl -fsSL https://cli.inference.sh | sh (and fetch binaries from dist.inference.sh), which fetches and executes remote installer code at runtime and is presented as a required dependency for the skill's research/fact‑checking workflow, so it is flagged.