Azure DevOps Orchestration

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection due to its core "Ship Workflow".
  • Ingestion points: Requirements, acceptance criteria, and reproduction steps are fetched from Azure DevOps work items via the az boards work-item show command (SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or instructions to treat work item content as untrusted.
  • Capability inventory: The agent has access to the Bash, Write, and Edit tools, as well as the az devops CLI, enabling it to modify the codebase, commit changes, and trigger pipelines (SKILL.md).
  • Sanitization: No sanitization or validation of the fetched work item data is performed before it is used to generate implementation plans and code.
  • [COMMAND_EXECUTION]: The orchestration logic involves extensive use of the Bash tool to perform git operations (git checkout, git pull, git push), interact with Azure services via the az CLI, and execute local test suites using auto-detected runners such as npm test or pytest (SKILL.md, references/ship-workflow.md).
  • [EXTERNAL_DOWNLOADS]: The skill configuration in SKILL.md and references/cross-plugin-patterns.md mentions several MCP servers and packages. These include @azure/mcp, @microsoft/azure-devops-mcp, @microsoft/powerbi-modeling-mcp, and @playwright/mcp, which are distributed by trusted organizations.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 03:34 AM
Security Audit — agent-trust-hub — Azure DevOps Orchestration