zap-fast

Warn

Audited by Snyk on Apr 6, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 1.00). This skill fetches and parses live data from multiple public KyberSwap endpoints (e.g., fast-zap.sh's fetch_zap_route call to https://zap-api.kyberswap.com, the token API at https://token-api.kyberswap.com, and the Earn Service/aggregator URLs referenced in SKILL.md) and directly uses those untrusted API responses to build and immediately broadcast transactions, so external content can materially change tool behavior.

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

  • Potentially malicious external URL detected (high risk: 1.00). The skill's fast-zap.sh calls the KyberSwap ZaaS endpoints at https://zap-api.kyberswap.com (and related Token API https://token-api.kyberswap.com) at runtime via curl to fetch a zap route and transaction calldata which are then directly used to build and immediately broadcast on‑chain transactions, so remote content both controls execution and is a required dependency.

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 build and immediately broadcast on-chain crypto "zap" transactions. It requires wallet configuration (keystore, env private key, Ledger/Trezor), exposes PRIVATE_KEY and wallet method env handling, calls a script (execute-zap.sh -> fast-zap.sh) that constructs and broadcasts a transaction to a ZapRouter address, verifies router addresses, checks allowances/balances, and returns txHash/blockNumber. The prompt repeatedly emphasizes irreversible, real-money on-chain execution with no confirmation. This is a direct crypto/blockchain financial execution capability (signing/sending transactions), not a generic tool.

Issues (3)

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
MEDIUM
Analyzed
Apr 6, 2026, 04:45 AM
Issues
3