rust-skill-creator

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

• Third-party content exposure detected (high risk: 1.00). The skill's inline and agent modes explicitly instruct the agent to fetch and extract documentation from public third-party sites (e.g., docs.rs and doc.rust-lang.org) and arbitrary user-provided URLs via agent-browser/WebFetch, causing the agent to ingest and act on untrusted third-party content when generating SKILL.md and reference files.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill issues runtime fetches (e.g., agent-browser open / WebFetch) to URLs like https://docs.rs/{crate}/latest/{crate}/ and https://doc.rust-lang.org/... and injects that fetched documentation into the model/skill-generation flow, meaning external content directly shapes prompts/instructions at runtime.