unlayer-image-editor
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch the integration script from 'https://cdn.unlayer.com/image-editor/embed.js' and the React component from '@unlayer/react-image-editor' on npm. Both are official distribution channels for the vendor 'unlayer'.
- [COMMAND_EXECUTION]: Provides standard development commands for checking package versions ('npm view') and installing dependencies ('npm install'), which are appropriate for the skill's purpose.
- [DATA_EXFILTRATION]: Code examples demonstrate typical image handling patterns, such as uploading an edited image blob to the user's own '/api/images' endpoint. There are no patterns suggesting the exfiltration of sensitive local data or credentials.
- [CREDENTIALS_UNSAFE]: The skill mentions a 'projectId' for AI features, but the examples use placeholder integers (e.g., '1234') rather than hardcoded secrets. It correctly instructs users to treat the project ID as a requirement for specific features without exposing sensitive tokens.
- [PROMPT_INJECTION]: The instructions focus strictly on technical implementation and configuration. No attempts to override agent behavior, bypass safety filters, or extract system prompts were found.
Audit Metadata