hyperliquid-review
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze Hyperliquid exchange records, creating a surface for external data ingestion. \n
- Ingestion points: Data enters the context via exchange records like
userFillsByTimeandorderStatusas specified inreferences/review-evidence.md. \n - Boundary markers: The workflow in
SKILL.mdand the structures inreferences/journal-contract.mdprovide boundaries for processing evidence. \n - Capability inventory: Capabilities are limited to local calculations and journaling; no network-write or arbitrary file-write tools are used. \n
- Sanitization: The bundled calculation script performs strict numeric validation and finite-number checks on all data passed to it. \n- [DYNAMIC_EXECUTION]: The skill executes a bundled Node.js script (
scripts/review.mjs) to perform deterministic trade accounting. The script is self-contained, does not use external dependencies, and is strictly limited to mathematical operations with validated inputs.
Audit Metadata