Skip to content

Scoring System — The Deterministic Quality Score (DQS)

Unmonitored documentation drift corrupts developer trust. A broken link degrades user experience, while an unredacted credential key requires immediate security incident response.

The Zenzic Deterministic Quality Score (DQS) provides a single 0–100 value computed from the concrete finding count across every check. Zero findings evaluates to 100/100. Given the same repository state and .zenzic.toml configuration, the score is 100% bit-for-bit deterministic across machines, operating systems, and runners.

DQS Category Weights

The Quality Score is a weighted composite of four distinct check categories:

Category Primary Commands Finding Codes Weight Bucket Cap
Structural Integrity zenzic check links Z101–Z105, Z107–Z109, Z113, Z121, Z124, Z410–Z411 30% 30 pts
Navigation Graph zenzic check orphans Z301–Z303, Z402 25% 25 pts
Brand & Governance zenzic check assets Z620, Z404–Z406, Z601, Z603, Z610–Z619 25% 25 pts
Content Excellence zenzic check all Z120, Z122, Z403, Z501–Z503, Z505–Z506, Z510–Z520 20% 20 pts

Inviolable Security Override

If any security finding is detected — Z201 Credential Scanner, Z202/Z203 Path Traversal Guard, Z204 Privacy Gate, Z205 XSS Gate — the Quality Score collapses to 0/100 unconditionally. A repository with active secret leaks receives zero quality credit.

Dual-Gate Architecture & Suppression Budget

The fail_under score threshold and suppression_cap operate as orthogonal, independently enforced quality constraints:

  • Score Gate (fail_under): Fails CI (Exit 1) if computed DQS score falls below the required threshold.
  • Governance Cap (suppression_cap): Fails CI (Exit 1) if active suppressions exceed the configured debt limit (default: 30).

Every active inline or per-file suppression deducts 1 Debt Point from the score:

\[\text{Max Achievable Score} = 100 - |F_s|\]

Worked Example & Mathematical Computation

Scenario: A repository contains 2 broken links (Z101), 3 orphan pages (Z402), 5 untagged code blocks (Z505), and 15 Z601 brand violations, with 8 active suppressions (cap = 30).

Stage Deduction Calculation Category Points Retained
Stage 1 — Security Gate No Z2xx findings detected Continue evaluation
Stage 2 — Structural 2 × 8.0 = 16.0 pts 14.0 / 30.0
Stage 2 — Navigation 3 × 4.0 = 12.0 pts 13.0 / 25.0
Stage 2 — Content 5 × 1.0 = 5.0 pts 15.0 / 20.0
Stage 2 — Governance 15 × 2.0 = 30.0 pts → amplified & capped to 25.0 0.0 / 25.0
\[\text{Subtotal} = 14.0 + 13.0 + 15.0 + 0.0 = 42.0\]

Applying active suppression debt (\(n = 8\)):

\[\text{Final DQS Score} = 42.0 - 8 = \mathbf{34 / 100}\]

CLI Quality Breakdown Ledger

Executing zenzic score prints a transparent breakdown ledger detailing raw deductions, category caps, and debt subtotals:

Terminal
✨ Quality Score: 65/100

╭─ Quality Breakdown ──────────────────────────────────────╮
│   Category     Issues  Weight  Raw Pts  Applied Pts      │
├──────────────────────────────────────────────────────────┤
│ ✓ structural      0      30%      0           0          │
│ ✓ navigation      0      25%      0           0          │
│ ✗ content         2      20%     -4          -4          │
│ ✗ brand          15      25%    -30         -25 (CAPPED) │
├──────────────────────────────────────────────────────────┤
│   Σ Subtotal                                71           │
╰──────────────────────────────────────────────────────────╯
  ! Technical Debt (6 suppressions)          -6 pts
  = Final Quality Score                      65 / 100

See Also