writing-guidelines

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

• Third-party content exposure detected (high risk: 0.85). This skill’s runtime workflow explicitly WebFetches public web content from an external URL (“Guidelines Source” at https://raw.githubusercontent.com/...), then feeds the fetched guideline text into the agent context to apply/check rules, which is outsider-authored free text.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs the agent to fetch runtime guidelines from https://raw.githubusercontent.com/vercel-labs/writing-guidelines/main/command.md and use that fetched content to control the rules/output format, so the external URL is fetched at runtime and directly controls agent behavior.