email-finder

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

• Third-party content exposure detected (high risk: 0.78). The skill’s runtime workflow includes fetching and parsing outsider-authored public web content (e.g., “curl + grep” on a personal site/portfolio URL, “curl” of GitHub commit .patch files, and “site:” dorking results), and those fetched pages/patches can contain arbitrary free text that may be ingested into the agent’s LLM context.