solana

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches account data from public Solana RPC endpoints (see SKILL.md RPC URLs and the fetch_account_data function in decode_account.py) and decodes untrusted on-chain/user-generated fields (including 'string' types), which the agent is expected to read and could materially influence subsequent actions.