A Bash tool to install the CodeQL CLI, pull in the official packs, build your Go project workspace, build a CodeQL database, run custom queries, and decode results. No bs—just repeatable steps and clear outputs.

☑️ Fighting CodeQL extensions in VS Code.
☑️ Wasting hours configuring analysis for each project.
☑️ Being limited to GUI tools when you live in tmux/vim/neovim.

• Zero-IDE Setup
  ./codeql_analyzer.sh install handles all dependencies
  (For those who are not VS Code friendly)

• Terminal-First Workflow
  Works with vim/emacs/nano + tmux/screen
  (Cloud/server-friendly for SSH workflows)

• Multi-Language Support : All Compiled languages (Java/C++ - when CodeQL's analysis works reliably in CLI mode)

  --language=javascript|cpp|java|python flags

1. Cloud servers with no GUI? No problem - runs headless
2. Need quick scans during CI/CD? Drops results in your pipeline
3. Hate IDE bloat? Your favorite terminal editor works fine

• Automated CodeQL CLI and pack installation
• Project structure scaffolding
• Go code database creation
• Static analysis with SARIF/BQRS outputs
• Automatic result decoding to text
• Unified commands for different workflows

• Bash (Linux environment)
• CodeQL CLI (auto-installed)
• CodeQL Queries (auto-installed)
• unzip, git, wget, golang (auto-installed if missing)
• Tested on Debian/Ubuntu systems

git clone https://github.com/madedis/CodeQL-Automation-Tool.git
cd codeql-analyzer
chmod +x main.sh

• Working on headless servers/cloud instances
• Needing quick scans without IDE configuration
• Running repeatable analyses across environments

sudo ./main.sh [OPTIONS] COMMAND [PARAMETERS]

sudo ./main.sh --help

• --work-dir DIR: Set working directory (default: current)
• --install-dir DIR: Set installation root (default: /opt/static_recon_codeql/workspace)
• --project-name NAME: Set project name (default: current directory name)
• --no-create : skip creating database if it's already created.

1. Install Dependencies

   sudo ./main.sh install

2. Create Project Structure

   sudo ./main.sh setup

3. Create Code Database

   sudo ./main.sh create-db /path/to/source-code

4. Run Analysis

   sudo ./main.sh analyze --queries-dir /path/to/queries --src-dir /path/to/source

   Run Analysis without the creation of a database again

   sudo ./main.sh analyze --queries-dir /path/to/queries --src-dir /path/to/source --no-create

5. Full Pipeline

   sudo ./main.sh full /path/to/source /path/to/queries

1. Install Core Components

   sudo ./main.sh install

2. Initialize Project

   sudo ./main.sh setup --project-name myapp

3. Build & Analyze

   sudo ./main.sh full ./src ./security-queries

📂 install-dir/
├── codeql-cli/          # CLI binaries
├── codeql-repo/         # Standard queries
└── [project-name]/
    ├── artifacts/       # Analysis artifacts(optional)
    ├── go-database/     # CodeQL database
    └── results/         # Analysis results (SARIF/TXT)

1. First run requires sudo for package installations
2. Default analysis format is SARIF v2.1.0
3. Logs stored in codeql_auto.log
4. Customize INSTALL_DIR in script for different locations

• Missing Dependencies: Ensure apt access and internet connection
• Database Errors: Delete corrupted go-database folder and retry
• Permission Issues: Run with sudo for system-wide installation

This is a time-saver - not a security silver bullet.

Test outputs before relying on critical systems.

Contributions welcome for bug fixes and QoL improvements.