composable-mini-apps
Pass
Audited by Gen Agent Trust Hub on Jul 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: This skill provides architectural guidelines for workspace organization and contains no executable code or malicious instructions.
- [DATA_EXFILTRATION]: The skill encourages security best practices by recommending that provider credentials and API details remain isolated within the specific app that owns the integration, rather than being shared or cloned across the workspace.
- [INDIRECT_PROMPT_INJECTION]: The skill describes a system of composable mini-apps that communicate via Agent-to-Agent (A2A) calls. This architecture introduces a surface for indirect prompt injection where an agent might process instructions from a sibling app. However, the guidelines mitigate this by recommending narrow prompts and structured artifact exchanges.
- Ingestion points: Sibling discovery blocks (discoverAgents) and A2A invocation responses (invokeAgent).
- Boundary markers: Not explicitly defined in this document.
- Capability inventory: The orchestration agent can call other apps and use provider-api-request.
- Sanitization: Not explicitly defined in this document.
Audit Metadata