feat: configurable health scoring weights via mcpwc.config.json

[himerus]

Summary

Allow Enterprises to Customize Health Scoring Weights

Currently health scoring uses hardcoded weights for each dimension (documentation, accessibility, naming, API consistency, etc.). Enterprise teams have different priorities — a design system team may weight accessibility at 3x while a rapid prototyping team may weight it at 0.5x.

Implementation

1. Add optional scoring section to mcpwc.config.json schema:

   {
     "scoring": {
       "weights": {
         "documentatio...
   

---

Created automatically by Automaker

Summary by CodeRabbit

• New Features

  • Configurable health scoring weights—users can now customize multipliers for individual health dimensions via configuration, with validation that rejects non-positive values.

• Documentation

  • Added comprehensive "Configurable Health Scoring Weights" section with JSON examples demonstrating accessibility-first and rapid-prototyping weight configurations.

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro

Run ID: ef52f3d6-7f49-4be3-9ca9-c9490722c1c8

📥 Commits

Reviewing files that changed from the base of the PR and between f46d56e and a39e02f.

📒 Files selected for processing (5)

• README.md
• packages/core/src/config.ts
• packages/core/src/handlers/dimensions.ts
• packages/core/src/handlers/health.ts
• tests/handlers/configurable-health-scoring.test.ts

---

Walkthrough

This PR adds configurable per-dimension health scoring weights by introducing a new optional scoring configuration section in the config file. It includes TypeScript types for the configuration, validation logic, a mapping of config keys to dimensions, and integration into the health scoring engine to apply weight multipliers from configuration.

Changes

Cohort / File(s)	Summary
Configuration Documentation README.md	Added documentation for the new scoring configuration object with scoring.weights per-dimension multiplier overrides, including behavior specification and JSON examples.
Configuration Types & Validation packages/core/src/config.ts	Added ScoringWeights and ScoringConfig exported interfaces; extended McpWcConfig with optional scoring field; introduced validateScoringWeights() and parseScoringConfig() helpers with filtering of non-positive values and warning output.
Dimension Weight Mapping packages/core/src/handlers/dimensions.ts	Added exported constant DIMENSION_WEIGHT_KEYS providing a read-only mapping from scoring config keys to dimension registry names.
Health Scoring Integration packages/core/src/handlers/health.ts	Added getEffectiveWeight() helper; updated scoreComponentMultiDimensional to read config weights once and apply multipliers to all dimension entries (native and external) via computed effectiveWeight.
Test Suite tests/handlers/configurable-health-scoring.test.ts	Comprehensive Vitest suite validating DIMENSION_WEIGHT_KEYS mappings, weight application behavior, config loading with validation, and scoring calculations with multiplier overrides.

Sequence Diagram

sequenceDiagram
    participant App as Application
    participant Loader as Config Loader
    participant Parser as Scoring Parser
    participant Scorer as Health Scorer
    participant Registry as Dimension Registry

    App->>Loader: loadConfig()
    Loader->>Parser: parseScoringConfig(raw.scoring)
    Parser->>Parser: validateScoringWeights()
    Note over Parser: Filter non-positive values,<br/>warn on invalid entries
    Parser-->>Loader: ScoringConfig object
    Loader-->>App: McpWcConfig with scoring

    App->>Scorer: scoreComponentMultiDimensional(component, config)
    Scorer->>Registry: DIMENSION_REGISTRY entries
    loop For each dimension
        Scorer->>Scorer: getEffectiveWeight(baseWeight,<br/>dimensionName,<br/>config.scoring.weights)
        Scorer->>Scorer: effectiveWeight = baseWeight × multiplier
    end
    Scorer-->>App: Scored dimensions with applied weights

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Possibly related PRs

• feat: CEM-Source Fidelity analyzer — cross-reference CEM declarations against actual source code #85: Adds a new dimension to the health scoring system; the changes in this PR enable configuration of weights for all dimensions including any new ones added through that PR.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feature/feat-configurable-health-scoring

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}