analyze-azdo-build

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local PowerShell scripts (Get-AzureDevOpsBuildAnalysis.ps1 and Retry-AzureDevOpsFailedStages.ps1) to perform build analysis and remediation tasks. It also utilizes the Azure CLI (az) and GitHub CLI (gh) for API interactions and build management.
  • [EXTERNAL_DOWNLOADS]: The skill fetches build metadata, timeline records, and task logs from Azure DevOps (dev.azure.com) and GitHub (github.com). These operations target well-known services and are strictly required for the skill's primary functionality of analyzing CI failures.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it processes external build logs and timeline records which could potentially contain adversarial instructions.
  • Ingestion points: Build timeline JSON files and task logs downloaded from Azure DevOps to the temporary directory (SKILL.md, scripts-reference.md).
  • Boundary markers: The instructions do not explicitly mandate the use of delimiters or warnings to ignore embedded instructions within the logs being analyzed.
  • Capability inventory: The skill has access to shell execution via PowerShell, network operations via CLI tools, and file system access to the project directory.
  • Sanitization: The underlying analysis script uses regex-based extraction to identify test failures and compilation errors, which provides a layer of data filtering before the agent processes the information.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 12:26 AM
Security Audit — agent-trust-hub — analyze-azdo-build