content-repurposer

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly requires ingesting and "Read entire source" material from arbitrary sources like newsletters, podcasts, articles and "archive content" (see SKILL.md "The Framework Fitting Process" and "Batch Processing Mode"), and those untrusted third‑party contents are parsed to extract snippets, determine tags/handles, and drive platform posting decisions—so external content can materially change agent behavior.