postiz-cli-agents

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

• Third-party content exposure detected (high risk: 0.70). The skill's core workflow (SKILL.md step 3: "postiz integrations:trigger … e.g. Reddit flairs, YouTube playlists") and the command reference show the agent invoking provider tools that fetch untrusted, user-generated content from public third‑party services via the Postiz API, which the agent is expected to read and use to decide settings or create posts—creating a clear vector for indirect prompt injection.