okx-cex-skill-mp

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the okx CLI and npx to perform marketplace management operations and install packages.
  • [REMOTE_CODE_EXECUTION]: The skill's primary function is to download and install third-party instructions and code from a remote marketplace. This behavior is mediated by security controls including Ed25519 digital signature verification and SHA-256 integrity checks to validate packages before installation.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes metadata (such as names and descriptions) from an external marketplace, which could potentially contain malicious instructions.
  • Ingestion points: Metadata processed during okx skill search and okx skill categories calls (SKILL.md).
  • Boundary markers: The documentation includes prominent warnings advising users to treat third-party skills with caution and review their contents before use.
  • Capability inventory: The skill has the ability to write to the file system, access the network via the CLI, and execute commands.
  • Sanitization: The platform verifies digital signatures and file hashes to ensure that the content has not been tampered with since being signed by the author.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:38 AM
Security Audit — agent-trust-hub — okx-cex-skill-mp