The Agent Skills Directory

Command Execution (Development Tools): The skill utilizes git for difference analysis and dotnet for running test suites. This execution of local binaries is the primary mechanism for generating the review and measuring test performance, aligning with standard developer workflows.
Dynamic Code Execution: During the review process, the skill triggers dotnet test to identify performance regressions, such as tests exceeding a one-second threshold. This runtime compilation and execution are necessary for providing the performance-based feedback described in the guidelines.
Indirect Prompt Injection Surface: As a code analysis tool, the skill reads and interprets file content that could potentially contain adversarial instructions. This is a characteristic inherent to LLM-based code review agents, and the risk is mitigated by the skill's intended use within a private development context on local files.

review-staged