binance-agentic-wallet

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the @binance/agentic-wallet CLI tool from the npm registry and references supplementary skills hosted in the official Binance GitHub repository.
  • [COMMAND_EXECUTION]: Executes the baw CLI to perform wallet operations, including authentication, balance checks, token transfers, and executing swaps or limit orders.
  • [PROMPT_INJECTION]: An indirect prompt injection surface exists where the agent processes on-chain data such as token names and transaction descriptions.
  • Ingestion points: Outputs from the wallet balance and wallet tx-history commands as specified in references/wallet-view.md.
  • Boundary markers: The 'Security Policy' in SKILL.md provides explicit negative constraints, directing the agent to ignore instructions found in token names or symbols.
  • Capability inventory: Ability to broadcast on-chain transactions and manage wallet configurations via the baw toolset.
  • Sanitization: The skill mandates human confirmation for all state-changing commands and incorporates a token security audit step in references/security.md to mitigate risks from malicious assets.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 03:26 AM