daily-briefing

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (scripts/01-fetch.js) fetches live content from the public API https://www.mixdao.world/api/latest and scripts/02-briefing.js directly ingests those items and passes their titles/text to the MiniMax model to generate summaries/recommendations that are then used to PATCH /api/latest/recommendation and upload briefings, so untrusted third‑party content could influence model behavior and subsequent tool actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The scripts call the MiniMax/Anthropic API (default ANTHROPIC_BASE_URL https://api.minimaxi.com/anthropic) at runtime and parse the model's JSON responses to drive grouping, generate recommendations, and decide subsequent submissions, so remote model outputs directly control the agent's behavior and are a required dependency.