pionex-trade

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the official @pionex/pionex-ai-kit package from the NPM registry during installation, which is a trusted vendor resource.
  • [SAFE]: Security best practices are integrated into the trading workflow, requiring the agent to use the --dry-run flag and obtain explicit user confirmation before finalizing any order placement or cancellation.
  • [SAFE]: Credential management is handled through a standard local configuration file (~/.pionex/config.toml) initialized by the vendor's onboarding tool, following industry standards for CLI tool security.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from CLI tool outputs in SKILL.md, representing a potential indirect prompt injection surface. The risk is mitigated by strict command-line parameterization and mandatory human-in-the-loop confirmation. 1. Ingestion points: Output from pionex-trade-cli commands. 2. Boundary markers: Explicit requirement for user confirmation. 3. Capability inventory: Spot order management (new, get, cancel). 4. Sanitization: Instructions to validate symbols and round values to exchange-required precision.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 04:38 PM
Security Audit — agent-trust-hub — pionex-trade