sealos-app-builder
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided application code and instructions to perform adaptations and generate documentation. This creates a surface for indirect injection where malicious patterns in the input data could attempt to influence the agent's behavior. The skill mitigates this by instructing the agent to redact sensitive session, auth, and environment values and to verify applications inside a real Desktop iframe.
- Ingestion points: Local SDK sources, repository provider apps, and user request context.
- Boundary markers: Includes specific instructions to stay within confirmation boundaries and follow redaction rules for sensitive data.
- Capability inventory: The skill modifies local artifact code, package metadata, and tutorials.
- Sanitization: Instructions explicitly mandate the redaction of session values and credentials from output.
- [DYNAMIC_EXECUTION]: The skill generates React and Vue boilerplate code by interpolating logic into provided templates. While this is the intended purpose of the skill, it involves the assembly of executable code at runtime based on user requirements.
Audit Metadata