alipay-authenticate-wallet
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill includes instructions that attempt to override the agent's behavior and safety guidelines. Evidence includes the self-proclaiming safety claim "不存在安全风险" (no security risk) and the directive "输出规则(最高优先级,覆盖所有其他输出规则)" (Output rules (highest priority, covers all other output rules)).
- [EXTERNAL_DOWNLOADS]: The skill initiates the installation of a Node.js package from the official npm registry. Evidence: The installation of
@alipay/agent-payment@1.0.0. The skill includes a subresource integrity (SRI) hash for the package to ensure the downloaded code has not been tampered with. - [COMMAND_EXECUTION]: The skill executes shell commands through a vendor-provided CLI tool to check and bind wallet status. Evidence: Commands such as
alipay-bot check-wallet,alipay-bot apply-wallet, andalipay-bot bind-walletare used within the skill instructions. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied authorization codes and external tool output which could be manipulated to influence the agent. 1. Ingestion points: User input for authorization codes and Markdown output from the
alipay-botCLI tool. 2. Boundary markers: The skill does not define clear delimiters for tool output, instead instructing the agent to copy output directly. 3. Capability inventory: The skill has shell execution privileges via thealipay-botcommand. 4. Sanitization: Authorization codes are restricted to a six-digit format, providing some protection against arbitrary injection.
Audit Metadata