Problem
The project follows a numpy-style commit prefix convention that is written down nowhere, and it has drifted into synonyms as a result. Across the last 120 commits: FIX: 12, RLS: 11, MAINT: 9, RFC: 7, DOC: 7, ENH: 3, PERF: 2, BUG: 2, TST:/TEST: 1 each, INFR:/IMP:/DOCS: 1 each.
FIX: and BUG: are the same thing; so are DOC:/DOCS: and TST:/TEST:; MAINT:, IMP: and INFR: overlap. A contributor has no way to pick correctly.
Worth aligning the vocabulary with the QEP-2 Type labels where they correspond — MAINT: ↔ maintenance, INFR: ↔ infrastructure, RFC: (or a new REF:) ↔ refactor — so a commit prefix and its issue's Type label tell the same story. That alignment is the main reason to bother.
Acceptance criteria
From the July 2026 technical-debt audit (AI-assisted; claims verified against 28d4b3b on 2026-07-25).
Problem
The project follows a numpy-style commit prefix convention that is written down nowhere, and it has drifted into synonyms as a result. Across the last 120 commits:
FIX:12,RLS:11,MAINT:9,RFC:7,DOC:7,ENH:3,PERF:2,BUG:2,TST:/TEST:1 each,INFR:/IMP:/DOCS:1 each.FIX:andBUG:are the same thing; so areDOC:/DOCS:andTST:/TEST:;MAINT:,IMP:andINFR:overlap. A contributor has no way to pick correctly.Worth aligning the vocabulary with the QEP-2 Type labels where they correspond —
MAINT:↔maintenance,INFR:↔infrastructure,RFC:(or a newREF:) ↔refactor— so a commit prefix and its issue's Type label tell the same story. That alignment is the main reason to bother.Acceptance criteria
CONTRIBUTING.mdor the contributing guideFrom the July 2026 technical-debt audit (AI-assisted; claims verified against
28d4b3bon 2026-07-25).