doc-drafting

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow instructs the agent to read comments on Google Docs (docs comments <id>), which can contain free text submitted by outside reviewers.

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

• Potentially malicious external URL detected (medium risk: 0.30). The installation guide instructs the user to execute npx skills add --global JarvusInnovations/agent-skills --skill doc-drafting, establishing a remote GitHub-backed operational dependency that fetches skill content from a third-party repository at runtime.