make-app-approval
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references product documentation hosted on Feishu (feishu.cn), which is a well-known enterprise collaboration and documentation service. This reference is used for retrieving generation rules and does not involve the execution of external code.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided functional plans to generate approval logic, which constitutes an indirect prompt injection surface. However, the skill mitigates this risk by establishing strict technical contracts (initiation-contract.md) and validation rules that prioritize schema-based logic over natural language keywords.
- [SAFE]: The instructions reinforce critical security practices, such as ensuring that approval-specific field permissions do not override general IAM policies and respecting system-level record locks (
qfei_is_locked) to prevent unauthorized edits during active approval processes. - [SAFE]: The skill explicitly defines boundaries that exclude high-risk operations like npm package installation, task management implementation, or administrative approval configuration.
Audit Metadata