dotnet-slopwatch

Detect LLM reward hacking by scanning C# source files, project files, and test files for patterns like disabled tests, warning suppressions, and empty catch blocks. This prevents LLM-introduced slop from masking real problems and degrading code quality.

Code slop in C# consists of LLM shortcuts like disabled tests, warning suppressions, and empty catch blocks that mask real problems. These shortcuts create hidden technical debt by bypassing actual error handling and suppressing necessary test failures.

Prevent disabled tests and empty catch blocks in CI/CD by integrating automated quality gates that analyze C# source files and test files. The scan identifies LLM-introduced slop and establishes a baseline to track new issues against existing technical debt.

Yes, you can track new warning suppressions by establishing a baseline that differentiates new issues from existing technical debt. This baseline approach ensures only LLM-introduced slop in recent code modifications is flagged during quality checks.

Yes, slopwatch analyzes C# project files and test files alongside source files. It scans these file types for LLM shortcuts such as disabled tests and warning suppressions, then reports any new issues introduced after LLM modifications.

Detection focuses on specific LLM slop patterns like disabled tests, warning suppressions, and empty catch blocks. It requires establishing an accurate baseline to differentiate new issues from existing technical debt, meaning pre-existing untracked issues may not be flagged.