AI Architecture & Code Intelligence Platform

TrueCourse provides AI-powered analysis for JavaScript, TypeScript, and Python codebases, detecting complex architectural violations and code quality issues often missed by traditional linters. It targets developers and architects seeking to improve code maintainability, reliability, and security by identifying problems like circular dependencies, layer violations, dead modules, and security anti-patterns, offering actionable fix suggestions.

How It Works

The platform employs a hybrid approach, combining fast, zero-cost tree-sitter static analysis for over 1,200 deterministic rules with LLM-powered semantic analysis for an additional 100 rules. This allows for deep inspection of code structure, quality, security, performance, and reliability, leveraging LLMs for more nuanced pattern detection and fix generation.

Quick Start & Requirements

• Primary install/run command: npx truecourse analyze (full analysis) or npx truecourse dashboard (web UI).
• Prerequisites: Node.js >= 20, Claude Code CLI on PATH (required for LLM rules).
• Setup: No explicit setup step; creates a .truecourse/ directory for configuration and results.
• Links: CLI commands and configuration options are detailed in the README.

Highlighted Details

• Ships with 1,200+ deterministic rules and 100 LLM rules across 8 categories (Security, Bugs, Architecture, Code Quality, Performance, Reliability, Database, Style).
• Integrates with Claude Code, offering conversational analysis and automated fixes via dedicated skills.
• Supports Git pre-commit hooks (truecourse hooks install) to automatically block commits introducing new violations, configurable by severity.
• Planned support for C#, Go, Rust, and PHP.

Maintenance & Community

For questions, feedback, or security reports, contact Mushegh Gevorgyan at mushegh@truecourse.dev. No specific community channels (e.g., Discord, Slack) or sponsorship details are listed.

Licensing & Compatibility

• License: MIT.
• Compatibility: The MIT license is permissive, generally allowing for commercial use and integration into closed-source projects.

Limitations & Caveats

LLM-powered rules necessitate a configured LLM provider via the Claude Code CLI. Pre-commit hooks can introduce significant latency (tens of seconds) on large codebases. Support for C#, Go, Rust, and PHP is planned but not yet implemented. Telemetry is enabled by default but can be opted out of.