Scoring

GHS uses a modular, tiered scoring system to measure repository health.

Modules

Module	Checks	Max Points	Weight (with lang module)	Weight (solo)
Core	40 scored + 3 INFO	74	60%	100%
.NET	20 scored + 3 INFO	34	40%	---

Point Values

Tier	Points	Description
Tier 1	4	Required --- fundamental quality
Tier 2	2	Recommended --- professional standards
Tier 3	1	Nice to Have --- polish and completeness

Core Module Maximum

Tier	Checks	Points Each	Subtotal
Tier 1	4	4	16
Tier 2	22	2	44
Tier 3	14	1	14
Total	40		74

.NET Module Maximum

Tier	Checks	Points Each	Subtotal
Tier 1	2	4	8
Tier 2	8	2	16
Tier 3	10	1	10
Total	20		34

Calculation

Scores are computed by querying GitHub Project items via jq pipelines — no local files or Python scripts are involved. See the GitHub Projects Format reference for the full jq queries.

Single module (core only)

score = earned_points / possible_points x 100

Multiple modules (core + language)

Each module is scored independently, then combined:

core_pct = core_earned / core_possible x 100
lang_pct = lang_earned / lang_possible x 100
combined = round(core_pct x 0.6 + lang_pct x 0.4)

All percentages rounded to the nearest integer.

Special Rules

WARN Exclusion

If a check returns WARN (cannot verify), it is excluded from both earned AND possible totals. This prevents penalizing repos for checks that can't be verified (e.g., permission issues). WARN items are never added to the GitHub Project, so they are automatically excluded from all jq scoring queries.

INFO Exclusion

INFO checks carry no points and are purely informational:

Core: Funding, Discussions Enabled, Commit Signoff

.NET: Target Framework, Package Count, Build System

Module Priority

When recommending next actions, core module items take priority over language module items at the same tier and point value.

Progress Bar

Scores are visualized with an 8-character progress bar:

Score: 52/74 (70%) ██████░░

Characters: filled with █ and empty with ░. Filled = round(percentage / 100 x 8).

Score Interpretation

Range	Meaning
90-100%	Excellent --- repo follows best practices
70-89%	Good --- some improvements available
50-69%	Fair --- notable gaps in repo quality
Below 50%	Needs attention --- fundamental issues