Grasp · Comprehension Index Scan Report

Who still understands openai/openai-python?

Computed from git metadata only; no source code contents were read.
touch-weighted · 3-year history window · last commit 58d ago · scores are comparable only against same-mode scans

Grasp Score

touch-weighted living knowledge across the repo

Active humans

8 / 104

contributors active in last 90 days vs. all-time

AI-attributed minimum

≥ 70.5%

hard evidence only. Just 79% of commits here carry any attribution signal, so the real share is higher

Inactive share

14.3%

work by authors with no commit in the window; silence, not proof of departure

Composition

79/10/7/4%

source / tests / docs / config share of scored work

Living, source only

24%

the headline recomputed over hand-written source alone

Score stability

stable

90-day-cliff variant reads 24%–25% across 60/90/180-day activity windows; read the width before the middle value

Dark-code ranking

Modules ranked by comprehension risk: the unknown share, amplified by ongoing change and thin ownership. The green meter is living knowledge.

Module	Risk	Living knowledge	AI-attr.	Inactive	Bus	Heat/180d
(root)	CRIT 92	8%	87%	10%	1	344
src/openai	CRIT 92	8%	88%	12%	1	130
src/openai/_utils	HIGH 76	5%	87%	15%	1	10
src/openai/resources	HIGH 75	18%	75%	13%	2	549
openai	HIGH 74	2%	0%	100%	0	0
tests	HIGH 69	25%	65%	20%	2	423
src/openai/cli	HIGH 68	9%	72%	70%	0	0
src/openai/types	HIGH 67	28%	69%	11%	2	2307
src/openai/lib	HIGH 66	30%	39%	46%	1	40
.github	HIGH 65	30%	49%	21%	2	60
examples	MED 57	30%	40%	50%	2	31
scripts	MED 56	40%	50%	0%	2	64
.inline-snapshot	MED 53	29%	0%	100%	0	0

One commit touched 67% of every file changed in this window. Wide, shallow commits, a formatter run or a license-header pass, add weight for their author across the whole surface while changing little. Worth looking at what that commit was before crediting the score to its author.

Public-repo caveat: "departed" includes drive-by contributors, overstating attrition relative to a company repo with a real roster. Engagement is a proxy for comprehension. Treat it as a potential indicator rather than a diagnosis.