bingx-spot-wallet
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input for API parameters including coins, network identifiers, and wallet addresses, which constitutes an indirect prompt injection surface.
- Ingestion points: User-supplied values for coin, network, address, amount, and addressTag extracted during the execution flow described in
SKILL.md. - Boundary markers: The agent is explicitly instructed to extract structured values from user intent and never copy raw text directly into parameters.
- Capability inventory: The skill can perform authenticated GET, POST, and DELETE requests to BingX API endpoints via a provided code template.
- Sanitization: Implements strict validation rules including regex patterns (e.g.,
^[A-Z0-9]{1,20}$for coins) and explicitly forbids the use of special characters like&,=,?,#, and newlines in parameters.
Audit Metadata