query-address-info

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill interacts with the user's environment by executing a local Node.js script scripts/cli.mjs. This script is zero-dependency and performs intended functionality only.
  • [EXTERNAL_DOWNLOADS]: The skill fetches real-time blockchain data from web3.binance.com and provides references to token logos at bin.bnbstatic.com. These represent official infrastructure from the skill author and are documented neutrally as intended functional requirements.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external token metadata such as names and symbols from API responses. The risk of indirect injection is addressed by strict regex-based validation of input parameters (wallet addresses and chain IDs) and the use of structured JSON output.
  • Ingestion points: Blockchain data retrieved from web3.binance.com within scripts/cli.mjs.
  • Boundary markers: The skill outputs structured JSON data to the agent context.
  • Capability inventory: Local script execution (node) and network access (fetch).
  • Sanitization: The script includes a dedicated validateAddress function that rejects whitespace, control characters, percent-encoded sequences, and non-ASCII characters.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 08:06 PM