Real-time, metric-based risk analysis for software projects — triggered automatically every time you save a file in VS Code.
What it does
Auto-FMEA analyzes your source code on every file save and computes a Risk Priority Number (RPN) for each class using three objective metrics: cyclomatic complexity, dependency impact radius (fan-in/fan-out), and test coverage detection. Results appear instantly as inline diagnostics and in a dedicated dashboard panel.
Supported languages
Python (.py) — powered by ast and radon
Java (.java) — powered by javalang
C# (.cs) — powered by custom static analysis
How it works
Auto-FMEA adapts the FMEA (Failure Mode and Effects Analysis) methodology — a standard used in aerospace, automotive, and safety-critical engineering — to software development:
Parameter
How it is measured
Severity
Fan-in: how many other classes depend on this class
Occurrence
Cyclomatic complexity of the class
Detection
Presence of a corresponding test file
RPN
Severity x Occurrence x Detection
RPN Range
Risk Level
>= 200
Critical
>= 100
High
>= 50
Medium
< 50
Low
Features
On-save trigger: analysis runs automatically when you save a .py, .java, or .cs file
Dependency analysis: detects which classes depend on the modified class, both within the same file and across the project
Duplicate class detection: flags classes with the same name defined in multiple locations
Actionable recommendations: at least 3 specific, metric-referenced recommendations per class
Dashboard panel: risk table, 5x5 risk matrix, project health score, and top risks — accessible via Ctrl+Shift+P > "Auto-FMEA: Show Dashboard"
