injective-trading-staking

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly fetches data from public Injective LCD REST API endpoints (e.g., "https://lcd.injective.network" and the GET /cosmos/... endpoints listed in SKILL.md) to ingest on-chain, user-generated validator/delegation/reward data which the agent reads and uses to determine staking/withdrawal actions, so untrusted third-party content could influence agent decisions.