rd-expense-check-free
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The SKILL.md file contains instructions for the user to install a full version of the tool by running
npx -y @alipay/agent-payment@latest install. This involves downloading and executing code from an external registry. - [COMMAND_EXECUTION]: The skill requires the
BashandReadtools to execute its local Node.js runner (scripts/run.mjs) and engine (scripts/engine/rd-expense-check.js). This is a benign and necessary capability for its stated purpose. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes external, potentially untrusted data.
- Ingestion points: Untrusted data enters the agent context via the
--inputargument inscripts/run.mjs, which reads a file or raw text. - Boundary markers: None. The input is processed line-by-line as raw text to be parsed into a table.
- Capability inventory: The skill has access to the
BashandReadtools, which could be leveraged if the agent follows instructions embedded within the processed data. - Sanitization: Absent. While the engine validates numerical formats, it does not sanitize or filter natural language content from the input rows.
Audit Metadata