codeql

What problem does it solve? Finding deep security vulnerabilities like SQL injection, command injection, and path traversal requires interprocedural data flow analysis that simple pattern matching cannot provide. This Skill automates the full CodeQL pipeline — building a database, modeling project-specific APIs with data extensions, and running filtered security queries — so you get trustworthy SARIF results instead of silent zero-finding runs. ## Core Features & Use Cases - Database Building with Quality Gates: Creates CodeQL databases for Python, JavaScript/TypeScript, Go, Java/Kotlin, C/C++, C#, Ruby, and Swift, with sequential build methods, macOS Apple Silicon workarounds, and quality assessment (baseline LoC, extractor errors). - Data Extension Generation: Detects custom wrappers around database calls, request parsing, and shell execution that CodeQL does not model, then generates source/sink/summary YAML models and validates them with before-and-after analysis. - Two Scan Modes: Run-all mode (security-and-quality plus security-experimental suites) and important-only mode (high-precision findings with post-analysis severity filtering), both using explicit .qls suite files to avoid silent query dropping. - Use Case: Point it at a repository and say "run a full CodeQL scan" — it discovers or builds the database, creates data extensions for unmodeled APIs, executes the analysis, and delivers filtered SARIF results in a single output directory. ## Quick Start Ask the assistant to run a CodeQL security scan on this repository and report the high-confidence vulnerabilities.