agent-zero-bridge

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly allows Agent Zero to invoke Clawdbot tools that can fetch external web content (see the "From Agent Zero → Clawdbot" web_search example in SKILL.md and the generic tool invocation implemented in scripts/clawdbot_client.js / scripts/lib/clawdbot_api.js), so untrusted public web/search results can be ingested and influence subsequent agent actions.