jb-agent-env

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

• Third-party content exposure detected (medium risk: 0.30). This skill reads outsider-managed free text at runtime only indirectly via the GitHub Gist registry/markdown contents (gh gist view ... --raw), which are user-editable and thus can contain arbitrary text that the agent will ingest when updating/processing registry entries.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill instructs the agent to fetch and follow content from the pinned Gist at https://gist.github.com/bjesuiter/98d5768dc360093affb8d8fdb064e45f (via gh gist view) at runtime, and that fetched markdown/registry directly controls installation instructions the agent must follow.