What problem does it solve? Manually chasing compiler errors, type mismatches, and security issues after every code change is slow and error-prone. This Skill runs an automated feedback loop that collects LSP diagnostics and AST-grep scan results after each edit, then iterates until errors reach zero or a defined completion condition is met. ## Core Features & Use Cases - LSP Diagnostic Feedback Loop: A PostToolUse hook captures language server errors and warnings after every Write/Edit operation across 16 supported languages. - AST-grep Security Scanning: Detects SQL injection, XSS, and hardcoded secrets using structural pattern rules during the loop. - Configurable Completion Conditions: Define zero-errors, zero-warnings, tests-passing, and coverage thresholds in ralph.yaml, with graceful degradation to linters when LSP servers are unavailable. - Use Case: While implementing a new authentication module, run the loop so every edit is checked by pyright or tsserver, errors are fixed iteratively, and the loop stops only when diagnostics are clean and tests pass. ## Quick Start Ask the AI to start the Ralph feedback loop with /moai:loop so it automatically fixes LSP errors until the code is clean.
