pr-to-video

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads a GitHub pull request via gh commands (fetch-pr.mjs and ingest.mjs), which ingest outsider-authored PR metadata, descriptions, comments, commit messages, and diffs into capture/pr.json and capture/diff.patch without prior explicit item selection by the agent.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the execution of a npx hyperframes skills update pr-to-video command which fetches remote skill material from an unverified public registry, constituting an unverifiable runtime external dependency of high provenance tier.