DRSource is an extensible, multi-language static analysis tool designed to detect vulnerabilities in source code. It uses a pluggable architecture to combine multiple detection techniques—from simple regex matching to advanced AST-based taint analysis—all driven by a central, user-configurable knowledge base.

It identifies security issues such as SQL Injection, Cross-Site Scripting (XSS), Command Injection, Hardcoded Secrets, and many others across all supported languages in a single, unified scan.

• Extensible Plugin Architecture The scanner automatically discovers and runs all available analyzer plugins. This allows new languages (e.g., Python, Go) and new analysis techniques (e.g., control-flow analysis) to be added without changing the corFeatures

• Centralized Knowledge Base All rules—from simple regex patterns to complex AST sources and sinks for taint analysis—are defined in a single, human-readable knowledge_base.yaml file. This makes it easy to add or customize rules for any supported language.

• Multi-Engine Analysis (in one scan) DRSource runs all plugins simultaneously, giving you a complete picture of your codebase:

  • Regex Engine: Uses a high-speed, general-purpose regex plugin to find known vulnerability patterns and hardcoded secrets in any file type.

  • AST Taint Engine: Performs deep data-flow analysis by parsing code into an Abstract Syntax Tree (AST) to track taint from user input sources (e.g., request.getParameter) to sensitive sinks (e.g., executeQuery).

• Persistent Database & Scan Comparison All findings are stored in a local SQLite database, allowing you to:

  • View scan history (--history).
  • Compare scans to find new, resolved, and persistent vulnerabilities (--compare).

• Robust CLI & Reporting The command-line interface offers powerful options, including:

  • Database initialization (--init-db).
  • Exporting results in multiple formats (--export [sarif|json|ascii]).
  • Taint-flow visualization (--show-trace).

Clone the repository and navigate to the project root:

git clone https://github.com/thesp0nge/dr_source.git
cd dr_source

Install the package (and all dependencies) in editable mode:

 pip install --editable .

This automatically registers all core plugins (JavaAstAnalyzer, RegexAnalyzer) so the scanner can find them.

Run dr_source against any source code directory. It will automatically detect all file types and run the appropriate analyzer plugins.

dr_source [OPTIONS] TARGET_PATH

• TARGET_PATH: The path of the codebase to analyze.
• --init-db: Initialize the database from scratch (drops and recreates tables).
• --history: Display the scan history for the project.
• --compare : Compare the latest scan with a previous scan specified by ID.
• --export [sarif|json|ascii]: Export scan results in the specified format.
• --output : Output file for the exported report.
• --show-trace: Display the full data-flow trace for AST-based vulnerabilities.
• --debug: Enable debug logging.
• --version: Show DRSource version and exit.

• Run a Standard Scan This one command runs all plugins (Regex, AST, etc.) on the codebase.

dr_source /path/to/my-project

• Export Results as SARIF (Ideal for uploading to GitHub Security)

dr_source --export sarif --output findings.sarif /path/to/my-project

• View Taint Traces

dr_source --show-trace /path/to/my-project

• Initialize the database

dr_source --init-db /path/to/my-project

Contributions are welcome! With the new architecture, there are two main ways to contribute:

1. Add/Improve a Rule:

• Simply edit the dr_source/config/knowledge_base.yaml file to add a new regex pattern or an AST sink/source.

2. Add a New Plugin:

• Create a new plugin package (e.g., dr_source/plugins/python/).
• Write your new analyzer class (e.g., PythonAstAnalyzer) that implements the AnalyzerPlugin API.
• Register your new plugin in the entry_points section of setup.py.

For all contributions, please fork the repository, create a new branch, and submit a pull request.

dr_source is licensed under the MIT License.

Special thanks to the maintainers of javalang for their work on Java AST parsing, which powers the Java taint analysis plugin.