The Agent Skills Directory

Command Execution: The skill utilizes several CLI tools including the Azure CLI (az), GitHub CLI (gh), and binlogtool to perform its tasks. These tools are used to fetch pipeline metadata, logs, and artifacts.
External Data Ingestion (Indirect Prompt Injection): The skill is designed to process external data sources such as Azure DevOps build logs and MSBuild .binlog files.
Ingestion points: External data enters the agent context via az devops invoke (logs) and az pipelines runs artifact download (artifacts).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content within these logs.
Capability inventory: The skill possesses capabilities to execute shell commands and perform file system operations (New-Item, Remove-Item).
Sanitization: No explicit sanitization of the fetched logs or artifact content is performed before processing.
File System Operations: The skill creates temporary directories within the user's $env:TEMP folder to store downloaded artifacts for analysis and includes automated cleanup steps to remove these files after the investigation is complete.
External Dependencies: The skill references a .NET global tool (binlogtool) available via NuGet. It follows best practices by instructing the agent to verify the tool's presence and provide setup instructions rather than attempting to install software automatically.

azdo-build-investigator