GRASP.
Grasp · Comprehension Index Scan Report

Who still understands sst/opencode?

Computed from git metadata only; no source code contents were read.
touch-weighted · full history, 3-year analysis window · last commit 56d ago · scores are comparable only against same-mode scans

Grasp Score
70
touch-weighted living knowledge across the repo
Active humans
63 / 1128
contributors active in last 90 days vs. all-time
AI-attributed minimum
≥ 6.4%
hard evidence only. Just 22% of commits here carry any attribution signal, so the real share is higher
Inactive share
16.4%
work by authors with no commit in the window; silence, not proof of departure
Composition
69/11/13/7%
source / tests / docs / config share of scored work
Living, source only
71%
the headline recomputed over hand-written source alone
Score stability
moderate
90-day-cliff variant reads 71%–86% 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.

ModuleRiskLiving knowledgeAI-attr. InactiveBusHeat/180d
nixHIGH 75
25%
65%29%1279
packages/cloudMED 51
32%
0%100%00
cmdLOW 46
39%
0%62%10
internalLOW 45
40%
0%61%10
packages/codemodeLOW 43
57%
29%0%1182
githubLOW 40
49%
4%66%16
appLOW 34
55%
0%88%10
packages/http-recorderLOW 32
66%
18%0%2166
(root)LOW 31
64%
6%34%3806
packages/sdkLOW 31
64%
12%17%5770
packages/statsLOW 30
70%
14%0%1787
sdksLOW 30
70%
0%21%1140
packages/webLOW 29
66%
13%16%33819
packages/llmLOW 29
71%
12%1%1671
infraLOW 29
66%
3%26%495
packages/consoleLOW 28
69%
3%25%22835
packages/coreLOW 28
70%
14%15%23363
packages/cliLOW 28
68%
18%6%3128
packages/functionLOW 28
72%
0%16%1160
packages/scriptLOW 26
66%
0%32%13

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.