integrating-jupiter
Warn
Audited by Snyk on Apr 19, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill's "Fresh Context Policy" explicitly requires fetching external documentation from dev.jup.ag at runtime (e.g., https://dev.jup.ag/docs/llms.txt and specific refs like https://dev.jup.ag/docs/swap/v2/order-and-execute.md), and those fetched docs are treated as authoritative guidance that can directly override/control the agent's instructions.
MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).
- Direct money access detected (high risk: 1.00). The skill is explicitly built for DeFi/crypto financial operations on Solana. It documents endpoints and flows that create, sign, and execute on-chain/value-moving transactions (e.g., Swap: GET /swap/v2/order -> sign -> POST /swap/v2/execute; Lend: POST /earn/deposit; Trigger: POST /orders/price with depositSignedTx; Recurring: POST /createOrder -> sign -> POST /execute; Send: POST /craft-send; Prediction markets: POST /orders). The quickstart includes a signAndSend helper that deserializes, signs, and sends VersionedTransaction to the RPC, and many playbooks describe creating and executing transactions that move tokens. These are specific, purpose-built APIs to initiate and complete monetary transfers/trades, so the skill grants direct financial execution capability.
Issues (2)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W009
MEDIUMDirect money access capability detected (payment gateways, crypto, banking).
Audit Metadata