aave-v3-plugin

Fail

Audited by Snyk on Apr 28, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E005: Suspicious download URL detected in skill instructions.

  • Suspicious download URL detected (high risk: 0.70). The links are direct downloads (raw .sh, .py and a GitHub release binary) from the okx GitHub organization — a legitimate, well-known vendor — but they include curl|sh and executable release artifacts which present a meaningful risk if the repository or release assets are compromised or if you run them without inspection.

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

  • Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and interprets external, untrusted data — e.g., SKILL.md and src/commands show it resolves Pool addresses and reads user account data via public RPC endpoints (plugin.yaml api_calls and rpc/get_pool / Pool.getUserAccountData) and calls onchainos/OKX defi APIs (defi_positions, defi_collect) whose values (health factor, available borrows, positions) directly influence warnings, dry-run vs execute decisions, and transaction submission.

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


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 for Aave V3: it constructs ABI calldata, performs ERC-20 approvals, and executes transactions (supply, withdraw, borrow, repay, set-collateral, set-emode, claim-rewards) via onchainos wallet contract-call and the aave-v3-plugin binary. It supports an execution flag (--confirm) to broadcast transactions from the user's wallet and returns txHashes. These are direct crypto/blockchain transaction capabilities (wallet signing and token transfers), not generic tooling.

Issues (4)

E005
CRITICAL

Suspicious download URL detected in skill instructions.

W011
MEDIUM

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

W012
MEDIUM

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

W009
MEDIUM

Direct money access capability detected (payment gateways, crypto, banking).

Audit Metadata
Risk Level
CRITICAL
Analyzed
Apr 28, 2026, 06:09 AM
Issues
4