autopredict

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly runs predict.py to scan live Polymarket markets (see SKILL.md and scripts/scan_markets.sh which note "Network access is required for live Polymarket reads"), so it ingests public, user-generated market/order-book data that the agent reads and uses to drive evaluations and trading decisions—exposing the agent to untrusted third-party content that could carry indirect prompt-injection-like instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's setup script performs a runtime git clone and subsequent install/run steps of the upstream repository at https://github.com/howdymary/autopredict.git, which fetches remote code that will be executed and is relied on by the skill.