uniapp-skill
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No patterns of prompt injection or attempts to override agent safety protocols were detected. The instructions include positive constraints to prevent the disclosure of tokens and privacy data.
- [DATA_EXFILTRATION]: No commands for unauthorized data access or exfiltration were found. The documentation of network APIs (uni.request) is limited to legitimate application development usage.
- [OBFUSCATION]: The skill does not contain hidden or malicious encoding. Base64 usage is confined to standard development examples for embedding small image assets.
- [REMOTE_CODE_EXECUTION]: No unauthorized remote execution patterns (such as curl-to-bash) were detected. Dynamic code capabilities are documented solely within the context of standard WebView communication features.
- [COMMAND_EXECUTION]: The skill references legitimate developer tools and CLI utilities (e.g., HBuilderX, miniprogram-ci) for building and deploying applications without any signs of arbitrary command injection.
Audit Metadata