ReWeaver AI DriftDetector

Product Introduction

1. Definition: ReWeaver AI DriftDetector is a deterministic, privacy-first code analysis tool that calculates a Production Drift Ratio (PDR). It falls under the technical categories of static application security testing (SAST), code quality analysis, and technical debt quantification.
2. Core Value Proposition: It exists to provide development teams with an objective, immediate, and granular measurement of how far a codebase is from being production-ready. Its primary value is in eliminating guesswork from code reviews and sprint planning by quantifying drift—the silent accumulation of technical debt and standards violations—across nine defined dimensions.

Main Features

1. Production Drift Ratio (PDR) Scoring: The core output is a single, actionable PDR score. This score is generated by applying a deterministic rule engine across nine production-readiness dimensions (e.g., security, performance, maintainability). The process involves cloning a public GitHub repository into ephemeral memory, scanning file contents without writing to disk, and applying predefined rules—no Large Language Models (LLMs) are used, ensuring consistency and zero token cost.
2. Line-by-Line Drift Attribution: Every finding is traced to a specific file and line number. This feature provides developers with precise, actionable intelligence, eliminating the need for manual searching. The scanner categorizes issues by severity and maps them to one of the nine core dimensions, enabling targeted remediation.
3. Historical Drift Analysis & Commit-Level Tracking: The tool performs a historical analysis by scoring every commit in the repository's history. This creates a visual drift timeline, showing precisely when and how the PDR score changed. Each data point represents an actual scan of a commit state, not an estimation, providing an accurate audit trail of codebase health over time.

Problems Solved

1. Pain Point: It addresses the "unknown unknown" problem in software development—code that appears to function but contains hidden issues that erode security, performance, and maintainability. It specifically solves the inefficiency of manual code review for detecting systemic drift and the opacity of technical debt.
2. Target Audience: Primary personas include Engineering Managers needing to quantify team output health, Lead Developers overseeing code quality, DevOps Engineers responsible for deployment readiness, and Security Engineers focused on proactive vulnerability prevention in CI/CD pipelines.
3. Use Cases: Essential scenarios include pre-merge validation for critical branches, auditing legacy or acquired codebases, establishing a quality baseline for new projects, and providing objective data for sprint planning and technical debt prioritization meetings.

Unique Advantages

1. Differentiation: Unlike subjective code review tools or LLM-based analyzers that provide probabilistic feedback, DriftDetector uses a deterministic rule engine for consistent, repeatable results. Unlike full CI/CD security scanners, it requires no integration, installation, or sign-up for public repos, offering instant analysis.
2. Key Innovation: Its privacy-by-design architecture is a key innovation. The scanner operates with a strict memory-only policy for code content, and for private scans, it uses short-lived, encrypted GitHub tokens with user-defined repository permissions. No scan results, code, or metadata are persisted, addressing a major trust barrier for code analysis tools.

Frequently Asked Questions (FAQ)

1. How does ReWeaver AI DriftDetector calculate the Production Drift Ratio? The DriftDetector calculates the PDR by applying a comprehensive set of deterministic rules across nine core dimensions of production readiness, such as error handling, security practices, and configuration management. Each violation contributes to the final score based on its severity, providing a single metric for code health.
2. Is my source code safe when using DriftDetector on a private repository? Yes. For private repos, the tool uses a secure OAuth flow to obtain a temporary, encrypted GitHub token with scopes you control. The code is streamed into volatile memory for analysis, never written to disk, and no scan results or code content are stored. Disconnecting the app or revoking the token on GitHub permanently severs access.
3. What are the nine dimensions of production readiness analyzed by DriftDetector? While the exact taxonomy is proprietary, the dimensions typically encompass areas like Security Vulnerability Detection, Performance Anti-Patterns, Configuration Consistency, Error Handling Robustness, Code Maintainability, Documentation Completeness, Dependency Hygiene, Testing Coverage Signals, and Deployment Readiness.
4. Can DriftDetector be integrated into a CI/CD pipeline? The current public version is a standalone web tool for on-demand analysis. However, the deterministic scanning engine and architecture are designed for automation, suggesting future potential for API or CI/CD integration to enforce drift thresholds automatically.
5. How does DriftDetector differ from using linters or static analysis tools like SonarQube? DriftDetector consolidates multiple linter-type checks into a unified, production-focused framework (the nine dimensions) and delivers a historical trend analysis out-of-the-box. It is designed for immediate, zero-configuration analysis rather than ongoing project integration, serving as a complementary audit tool.