Specification › Coverage map

AI Contributor Specification — Coverage map

Coverage

Coverage map

Every rule in the spec, aggregated by weight, pillar, and minimum level. Numbers are derived from the rule catalog at build time — they cannot drift.

Rules240

MUST54

MUST·when68

SHOULD78

MAY40

By pillar

Pillar	Clauses	Rules	MUST	MUST·when	SHOULD	MAY
Engineering Foundation§1–4	4	33	19	0	10	4
Security§5–9	5	38	7	11	12	8
Quality & Reliability§10–14	5	33	5	8	14	6
Release§15–16	2	17	4	4	6	3
AI Agents§17–20	4	47	5	18	15	9
AI Risk§21–22	2	28	2	16	7	3
Oversight§23–29	7	44	12	11	14	7

By level (cumulative)

Level	Rules at this level	Cumulative	Closure
L0	27	27	11%
L1	114	141	59%
L2	81	222	93%
L3	18	240	100%
L4	0	240	100%

This generated projection is rendered from tools/spec-authoring/templates/AI-CONTRIBUTOR-COVERAGE.md.template plus AI-CONTRIBUTOR-RULE-CATALOG.json. The template owns explanatory prose and placement; the catalog owns row grouping, scope, level, pillar, and display metadata. Regenerate it with tools/spec-authoring/generate-coverage.ts; do not edit the generated output by hand.

How to read the numbers

Use this file for planning, not for scoring. It tells you how many checklist rows exist at each level and pillar so you can estimate effort before an audit.

For a conformance claim, use AI-CONTRIBUTOR-CHECKLIST.md. If this file and the checklist disagree, rerun the catalog-backed generators; both files are projections of AI-CONTRIBUTOR-RULE-CATALOG.json.

At a glance

207 total rows
34 unconditional MUST + 57 MUST when applicable + 76 SHOULD + 40 MAY
6 rows at L0 — Baseline Hygiene; 40 rows at L1 — Hardened; 28 rows at L2 — AI Assisted; 14 rows at L3 — AI Authored; 79 rows at L4 — AI Autonomous
40 optional MAY rows

By scope

Scope	Rows
`MUST`	34
`MUST when applicable`	57
`SHOULD`	76
`MAY`	40
Total	207

By pillar

Pillar	Total	`MUST`	`MwA`	`SHOULD`	`MAY`
1 · 🏗️ Engineering Foundation	26	11	1	10	4
2 · 🛡️ Security	34	5	9	12	8
3 · 🎯 Quality & Reliability	29	3	6	14	6
4 · 🚀 Release	13	2	3	5	3
5 · 🤖 AI Agents	39	3	13	14	9
6 · ⚠️ AI Risk	26	2	14	7	3
7 · 🧭 Oversight	40	8	11	14	7

By minimum conformance level

Level	Rows	`MUST`	`MwA`	`SHOULD`
L0 — Baseline Hygiene	6	5	1	0
L1 — Hardened	40	22	18	0
L2 — AI Assisted	28	7	21	0
L3 — AI Authored	14	0	14	0
L4 — AI Autonomous	79	0	3	76
— (`MAY` — never required)	40	—	—	—

Cumulative closure bar

This table shows how many MUST and MUST when applicable rows a repository must close, or mark Not relevant with evidence, to claim each level. The “worst case” column assumes every MUST when applicable trigger applies. Many repositories will trigger fewer rows.

Claiming level	Cumulative `MUST` + `MwA` (worst case)	New rows over previous level
L0	6	—
L1	46	+40
L2	74	+28
L3	88	+14
L4	91 (plus 76 `SHOULD` rows resolved)	+3 `MUST`/`MwA`, +76 `SHOULD`