aisa-tavily-hardened

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

• Third-party content exposure detected (high risk: 1.00). The skill's core scripts (scripts/search.mjs and scripts/extract.mjs) call AIsa's Tavily endpoints (https://api.aisa.one/apis/v1/tavily/search and .../extract) to fetch/search and return raw content from arbitrary public URLs and web search results, which the agent is expected to read and could materially influence its next actions—exposing it to untrusted third‑party content that could enable indirect prompt injection.