bingx-swap-market
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes market data (such as klines, order book, and ticker data) retrieved from external BingX API endpoints. This creates an attack surface where a compromised or malicious API response could attempt to inject instructions into the agent's context.
- Ingestion points: Data is fetched from
open-api.bingx.comand fallback domains (e.g.,bingx.pro) via thefetchSignedfunction. - Boundary markers: The skill documentation provides rules for parameter extraction but does not define explicit delimiters to isolate API-provided content from the agent's internal instructions.
- Capability inventory: The skill has network access via the
fetchAPI and utilizes HMAC signing with provided credentials. - Sanitization: A
validateParamsfunction is implemented to check for and reject special characters (&,=,?,#,\r,\n) in trading symbols and other parameters, mitigating direct injection into the query string. - [EXTERNAL_DOWNLOADS]: The skill requires the
json-bigintNode.js package to accurately parse large integers returned in API responses, which is a standard requirement for financial data processing.
Audit Metadata