clipping

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly requires fetching and extracting content from arbitrary URLs (see "Navigate URL", "Read first 4k chars", "Extract full Markdown" and the hermes-clip --url usage), and uses that untrusted, third-party content to decide categorization, tags, and merge/create actions, so external content could influence agent behavior.