The Agent Skills Directory

Command Execution: The skill utilizes standard .NET development tools, such as dotnet test and msbuild, to collect code coverage and metrics. These operations are performed using repository-compatible commands and are scoped to the intended purpose of code analysis.
Indirect Prompt Injection Surface: The skill processes external data, including source code and Cobertura XML reports, which could theoretically contain malicious instructions. However, the skill includes defensive measures to mitigate this risk.
Ingestion points: Source code files and Cobertura XML reports are read during the analysis workflow (SKILL.md).
Boundary markers: The instructions explicitly direct the agent to ignore source code comments when calculating complexity and to verify the integrity of XML reports, providing a clear boundary between data and instructions.
Capability inventory: The skill has the capability to read local files and execute shell commands (dotnet test, msbuild) within the development environment.
Sanitization: The skill relies on structured XML parsing and specific code construct counting rather than executing or interpreting free-text content from the analyzed files.

crap-score