skills/okx/agent-skills/earn-hunter/Gen Agent Trust Hub

earn-hunter

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEPERSISTENCECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PERSISTENCE]: The skill establishes automated monitoring by configuring the system crontab and creating a macOS LaunchAgent (com.okx.earn-hunter.plist). These mechanisms ensure the scan script runs at defined intervals independently of the agent session.
  • [COMMAND_EXECUTION]: The skill executes local shell scripts (scan.sh) and the okx-trade-cli to retrieve market data, check account balances, and facilitate earn subscriptions. It also uses chmod to set execution permissions on its distributed scripts.
  • [EXTERNAL_DOWNLOADS]: During activation, the skill installs the @okx_ai/okx-trade-cli package from the NPM registry. This is a vendor-owned resource used for interacting with OKX services.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a vulnerability surface due to its data processing flow. 1. Ingestion points: JSON data returned from the OKX Flash Earn and Savings APIs. 2. Boundary markers: No explicit boundary markers are used in the notification templates. 3. Capability inventory: The skill can execute shell commands, write state files, and perform network requests via curl. 4. Sanitization: The scan script utilizes jq to robustly parse and filter the API responses before they are presented to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:38 AM
Security Audit — agent-trust-hub — earn-hunter