helius-jupiter
Warn
Audited by Snyk on May 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and use third‑party token metadata and off‑chain content (e.g., "Includes off‑chain metadata (Arweave, IPFS)" in references/helius-das.md) and SKILL.md/integration-patterns require using DAS and Jupiter token/price APIs to build token selectors, enrich metadata, and drive swap/transaction decisions—so the agent will ingest untrusted, user‑generated content from public sources that can materially influence actions.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs loading and executing third‑party JavaScript at runtime via the swap widget script URL https://plugin.jup.ag/plugin-v1.js, which causes remote code to run in the app and is presented as a required dependency for the Jupiter Plugin swap widget.
MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).
- Direct money access detected (high risk: 1.00). The skill is explicitly designed to perform on-chain financial operations. It references Jupiter Swap APIs (token swaps and /execute calls), Jupiter Trigger/Recurring APIs (limit orders, DCA, order management), Jupiter Lend write SDKs (deposit/withdraw/borrow), Perps (open long/short positions), and Helius Sender for direct transaction submission (with explicit rules: skipPreflight, fee tips, use of getPriorityFeeEstimate). It instructs how to build and submit trades, recurring buys, limit orders, lend/borrow transactions, and optimized transaction submission—all of which move crypto assets on Solana. These are specific crypto financial execution capabilities (wallet/transaction submission, swaps, order placement), so this skill grants direct financial execution authority.
Issues (3)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W009
MEDIUMDirect money access capability detected (payment gateways, crypto, banking).
Audit Metadata