meteora

Fail

Audited by Snyk on Aug 4, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 0.80). The skill contains code examples that assign an API key string (JUPITER_API_KEY = 'your-api-key') and a literal secret key placeholder (Keypair.fromSecretKey(/* your secret key */)), which encourages embedding secrets verbatim in generated code/commands and therefore could require the LLM to handle/output secret values directly.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

  • Direct money access detected (high risk: 1.00). The SKILL.md describes explicit DeFi SDK methods that create and send on-chain financial transactions (e.g., swap(), buy(), sell(), deposit(), withdraw(), fundReward(), stake(), zap.*) and shows examples of constructing/signing/sending transactions (sendAndConfirmTransaction, Keypair.fromSecretKey, Wallet). These are specific financial execution operations (swaps, liquidity deposits/withdrawals, staking, presale deposits, pool creation, etc.), not generic tooling — therefore the skill grants direct financial execution capability.

Issues (2)

W007
HIGH

Insecure credential handling detected in skill instructions.

W009
MEDIUM

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

Audit Metadata
Risk Level
HIGH
Analyzed
Aug 4, 2026, 12:31 AM
Issues
2
Security Audit — snyk — meteora