aspire-integration-testing
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes how to integrate AI assistants with application telemetry (logs and traces) using the Model Context Protocol (MCP).
- Ingestion points: AI assistants query the running application's state, console logs, and distributed traces via the Aspire MCP server (described in
ci-and-tooling.md). - Boundary markers: The skill does not include specific instructions or markers to delimit application-generated telemetry from agent instructions.
- Capability inventory: When connected via MCP, the agent can access real-time console output, resource states, and telemetry data across the distributed application.
- Sanitization: There is no mention of sanitizing or escaping application logs that might contain untrusted user-supplied data before they are queried by the AI agent.
- [SAFE]: The skill recommends setting
ASPIRE_ALLOW_UNSECURED_TRANSPORT=truefor debugging purposes. While this reduces transport layer security for local development, it is a documented and standard configuration for the .NET Aspire framework. - [SAFE]: The skill documentation includes commands for installing the official
aspire.clitool viadotnet tool install. These are standard development operations for the Microsoft/Aspire ecosystem.
Audit Metadata