grimoire-hyperliquid
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes the
@grimoirelabs/clipackage, which can be executed dynamically vianpxfrom the NPM registry. - [COMMAND_EXECUTION]: The agent is instructed to execute shell commands using the
grimoireCLI to perform market queries and initiate withdrawals. - [INDIRECT_PROMPT_INJECTION]: The skill ingests market data (order books and metadata) which represents a potential vulnerability surface if the external data source were to be compromised.
- Ingestion points: Market data retrieved through commands like
l2-bookandmetaas described inSKILL.md. - Boundary markers: None identified in the instruction text.
- Capability inventory: The skill uses CLI tools to perform network requests and local file system access for keystores in
SKILL.md. - Sanitization: No specific sanitization or validation logic is detailed for the data returned from the CLI tools.
Audit Metadata