watching-authored-prs

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

• Third-party content exposure detected (medium risk: 0.30). The workflow monitors and consumes GitHub pull request comments and review summaries via GraphQL queries in shared/pull-request-comments.md, which ingest outsider-authored free text without prior selection.

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

• Potentially malicious external URL detected (medium risk: 0.30). The instruction references an external skill via npx skills@latest add bostonaholic/skills --skill addressing-pr-comments, which downloads executable content from an external repository locator (bostonaholic/skills), constituting an unverifiable external runtime dependency with common/reputable unofficial provenance.