“Which testing methodologies are actually used by AI code-reviewers and AI-augmented testing systems like Gerrit, CodeRabbit, GitHub Copilot PR Review, Cursor, Replit Agents, Qodo, etc.”

The key is that none of these platforms use all methodologies.

Each one specializes in a subset: static analysis, semantic reasoning, test generation, and workflow QA.

Below is a structured map so you can see who uses what.

High-Level View

Most AI code-testers fall into three functional categories:

1. Static Reviewers (Gerrit, GitHub PR Review)
2. AI PR Review + Test Generation (CodeRabbit, Qodo, Cursor AI)
3. Full Agentic Testers (Replit Agents, Claude Code, some Qodo modes)

1. Gerrit (Traditional + AI Plugins)

Gerrit is fundamentally a code review and static analysis platform.

Its AI capabilities depend on plugins (e.g., Gerrit AI, Sonar, CodeAI integrations).

Core testing methodologies Gerrit supports:

• Static Analysis
• Security Analysis (SAST)
• Style/Linting Checks
• Dependency Vulnerability Scans
• Regression Risk Estimation (AI plugin-based)