crypto-trading-advisor

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly instructs the agent to consult public third‑party sources (e.g., Alternative.me, DefiLlama, Coinglass, Glassnode, Whale Alert, Arkham, etc.) for market/sentiment/on‑chain data and the README/example explicitly shows Claude "fetches fear/greed index, funding rates, on‑chain data, news..." so the agent is expected to ingest and act on untrusted web content that can change trading decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The package's installer (bin/index.js) fetches the skill file at runtime from https://raw.githubusercontent.com/0xrikt/crypto-skills/main/crypto-trading-advisor/SKILL.md, and that fetched SKILL.md directly contains the agent's instructions/prompts and is required for the automated installation—so the URL is a runtime dependency that controls agent prompts.