retro-session

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

• Third-party content exposure detected (medium risk: 0.30). The workflow scans local agent session transcript files (.jsonl), which can contain tool outputs, messages, and command results written by anyone during an agent session, serving as an untrusted input source.

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

• Potentially malicious external URL detected (medium risk: 0.30). The README instructs running npx skills add matthewzammit/skills --skill retro-session, which fetches and installs an external skill package from GitHub at runtime.