exa-search

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path sends the user’s query/payload to Exa and then returns Exa’s synthesized context/output.content back into the agent, which is outsider-authored free text fetched from the public web via Exa’s /search//contents//findSimilar//answer endpoints.