grimoire-hyperliquid

Warn

Audited by Snyk on Feb 16, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.70). The skill queries public Hyperliquid venue data (mids, l2-book, open-orders, meta/spot-meta) via the Grimoire CLI and can emit those third‑party market/metadata snapshots (including --format spell) which the agent reads, so it ingests untrusted public content from the Hyperliquid venue.

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

  • Direct money access detected (high risk: 1.00). The skill explicitly exposes a withdraw command for Hyperliquid with parameters to send funds: "--amount --keystore [--password-env ] [--destination ]". The notes state "withdraw is stateful and requires explicit user confirmation plus keystore credentials." This is a specific, built-in capability to move money/crypto (initiate withdrawals), not a generic tool. Other commands are read-only, but the presence of the withdraw operation with keystore and destination makes this direct financial execution.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 09:45 AM