analyze-code

name	analyze-code
description	Run dart_code_linter analysis on the project and explain results Use when this capability is needed.
metadata	{"author":"bancolombia"}

Analyze Code Quality

You are an AI assistant helping a developer analyze their Dart/Flutter project using Dart Code Linter (DCL).

Steps

1. Check if DCL is installed by looking at pubspec.yaml for dart_code_linter in dev_dependencies. If it is not installed, instruct the user to run:

   dart pub add --dev dart_code_linter
   

2. Check for configuration in analysis_options.yaml. If there is no dart_code_linter section, suggest a basic configuration with recommended rules and metrics.

3. Run the analysis using the CLI:

   dart run dart_code_linter:metrics analyze lib --reporter=json
   

4. Interpret the results and present a summary organized by severity:

   • Critical: High cyclomatic complexity (>20), very long methods (>100 SLOC), deep nesting (>5 levels)
   • Warning: Moderate metric violations, anti-pattern detections (long-method, long-parameter-list)
   • Info: Style rule violations, minor code quality suggestions

5. Highlight the top issues to address first, prioritized by:

   • Files/methods with the highest cyclomatic complexity
   • Methods with the most parameters
   • Files with the most rule violations

6. Suggest actionable fixes for each critical issue:

   • For high complexity: recommend extracting methods or simplifying conditionals
   • For long methods: identify logical sections that could become separate functions
   • For deep nesting: suggest guard clauses or early returns
   • For rule violations: explain the rule and show the corrected code

7. Provide a summary with:

   • Total number of files analyzed
   • Count of violations by severity
   • Top 3 files that need the most attention
   • Overall code health assessment

---

Source: bancolombia/dart-code-linter — distributed by TomeVault.