run-toolset-tests
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill automates several shell commands using
gitand theaz(Azure CLI). These commands are used to identify the current branch state, push code to a remote repository, and trigger build pipelines. These are standard operations for a developer-oriented automation tool within a CI/CD context. - Interaction with Remote Resources: The skill facilitates communication with external platforms including GitHub (
github.com/dotnet/roslyn) and Azure DevOps (dev.azure.com/dnceng). These interactions are directed towards official repositories and internal engineering environments of a well-known organization, representing legitimate operational behavior for its stated purpose. - Indirect Prompt Injection Surface: The skill ingests data from the local environment, such as the current git branch name and commit SHA, and interpolates this data into shell commands and API requests. While this creates a potential surface for indirect injection if local metadata is maliciously crafted, the risk is minimized by the skill's specific use case and the requirement for authenticated access to the target infrastructure.
Audit Metadata