second-brain

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

• Third-party content exposure detected (high risk: 0.90). The skill's core workflow (SKILL.md and references/ingest_sources.md) and its /ingest and raw-folder compile commands explicitly fetch and parse untrusted public sources — e.g., Twitter/X via xurl, LinkedIn and arbitrary web URLs via defuddle/agent-browser, YouTube transcripts via yt-dlp — and the agent is instructed to read those captures and incorporate them into the maintained wiki and to drive planning, execution, and automated outputs, so third‑party content can materially influence agent behavior.