migrate-mstest-v1v2-to-v3
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Indirect Prompt Injection: The skill analyzes project files and source code as part of its migration workflow, which presents a surface for indirect prompt injection if those files contain adversarial instructions intended to influence the agent. * Ingestion points: The agent reads project files (.csproj, .sln) and source code in Step 1 and Step 5. * Boundary markers: The instructions do not provide explicit delimiters or instructions to treat ingested code as untrusted data. * Capability inventory: The skill utilizes file modification capabilities and executes shell commands such as 'dotnet build' and 'dotnet test' (Steps 5 and 7). * Sanitization: No sanitization or validation of the project content is specified before processing.
- Command Execution: The agent is instructed to run build and test commands to verify migration success. While standard for development tasks, these commands execute code within the project, making it important to only process projects from trusted sources.
Audit Metadata