repowiserepowise
Sign in
Mnehmos/mnehmos.oculus.webapp
OverviewDocsArchitectureKnowledge GraphFilesCode HealthRefactoring

People & History

CommitsContributorsDecisions
ChatPro
Stats
repowiserepowise
ExplorePricingDocs
Sign inIndex repoIndex your repo free
repowiseMnehmos/mnehmos.oculus.webapp

Commits

Every commit scored for change-risk against this repo's own history, so 'elevated' means elevated here rather than on some global curve.

Needs review

12of 38 scored

12 commits sit in this repo's top risk tercile, which is 32% of the 38scored. The cut is drawn against this codebase's own history rather than a global curve, so a quiet repo still fills its top band, and here it starts at 8.7 out of 10. What pushes a commit up is size and spread together: a large change confined to one area scores below a smaller one scattered across a dozen files.

Fix commits411%Commits whose subject reads as a bug fix rather than new work.Written by an agent3695%Read from commit trailers, so it counts what was declared.Change diffusion0.76bitsShannon entropy of a commit's churn across its files. Zero is a single file, and every extra bit is a doubling of how widely the change spread.Review threshold8.7out of 10Score a commit has to clear to land in this repo's top tercile.

How the work changed shape

Commit categories over time, read off the subject line. Fixes carry the accent because that is the series this chart exists to show.

How this repo's commit mix shifts over time.

Feature58%
Fix13%
Refactor3%
Docs8%
Test11%
Chore5%
Other3%
95%of indexed commits are agent-attributed(36 of 38)
claude (36)

Review-priority queue

Ranked by change-risk, highest first. Priority is a tercile of this repo's own distribution, so a quiet repo still fills its top band.

Commit review-priority queue
#CommitAuthorWhenLinesRiskTop driver
1
5c4d932ffeat(beta): initial v0.1 release - gaze-tracking pedagogy instrument with brick content model, typed event detection, live telemetry, and two starter lessons (gravity, recursion)
Mnehmos
3mo ago+2.8K -0
99%Elevated
more lines added than baseline
2
6a089b35feat(ensemble): multi-head classifier — train multiple versions at once
Mnehmosclaudeassisted
3mo ago+542 -389
95%Elevated
more lines added than baseline
3
6bd0c6c0feat(gaze-mode): add regression mode — predict viewport coords, hit-test
Mnehmosclaudeassisted
3mo ago+420 -210
95%Elevated
more lines added than baseline
4
be4ea95dPivot gaze pipeline to classification-first
Mnehmos
3mo ago+355 -152
89%Elevated
more lines added than baseline
5
8a71a066feat(calibration): replace 9-point with per-brick classifier calibration
Mnehmosclaudeassisted
3mo ago+420 -91
89%Elevated
more lines added than baseline
6
c7a53dc6test(fixtures): synthetic landmark generator + DOM test harness
Mnehmosclaudeassisted
3mo ago+273 -0
84%Elevated
more lines added than baseline
7
f01ec944feat(features): geometric feature extraction for gaze classifier
Mnehmosclaudeassisted
3mo ago+297 -0
84%Elevated
more lines added than baseline
8
1e9fd92ffeat(reader-cursor): mouse-driven reading + word-level tracking
Mnehmosclaudeassisted
3mo ago+347 -5
80%Elevated
more lines added than baseline
9
2165909efeat(classifier): TensorFlow.js MLP for feature→brick classification
Mnehmosclaudeassisted
3mo ago+226 -0
78%Elevated
more lines added than baseline
10
8c5ef2f9feat(calibration): grid method — viewport-fixed dots, no scroll
Mnehmosclaudeassisted
3mo ago+193 -9
72%Elevated
more lines added than baseline
11
6e9f2be3docs: update PROJECT_KNOWLEDGE.md for v0.2
Mnehmosclaudeassisted
3mo ago+189 -46
72%Elevated
more lines added than baseline
12
9cf74c79feat(face-landmarker): add MediaPipe Face Landmarker wrapper
Mnehmosclaudeassisted
3mo ago+189 -0
72%Elevated
more lines added than baseline
13
0bad03e5feat(export): LLM-grade reading trace — word-level data + event sourcing
Mnehmosclaudeassisted
3mo ago+160 -27
64%Typical
more lines added than baseline
14
2172a91dfeat(pagination): page field on bricks, no auto-scroll, tighter classifier gates
Mnehmosclaudeassisted
3mo ago+234 -60
64%Typical
more lines added than baseline
15
f8ca2e9adocs(handoff): add VALIDATION_HANDOFF.md for Vario
Mnehmosclaudeassisted
3mo ago+182 -0
64%Typical
more lines added than baseline
16
7072dd8dfeat(gaze): rewrite pipeline to consume classifier output
Mnehmosclaudeassisted
3mo ago+140 -35
59%Typical
more lines added than baseline
17
9182c1e9test(pipeline): integration test for Features → Classifier → Gaze
Mnehmosclaudeassisted
3mo ago+135 -0
57%Typical
more lines added than baseline
18
f405966ffeat(events): dwell-gated first_read + mouse-last-word + gaze override
Mnehmosclaudeassisted
3mo ago+108 -10
54%Typical
more lines added than baseline
19
d3341778test(features): unit tests for feature extraction + normalization
Mnehmosclaudeassisted
3mo ago+115 -0
51%Typical
more lines added than baseline
20
c8a90347test(classifier): unit tests for MLP build/train/predict/argmax
Mnehmosclaudeassisted
3mo ago+97 -0
47%Typical
more lines added than baseline
21
7e5f6fdcfeat(deps): remove WebGazer, add TensorFlow.js CDN + v0.2 module slots
Mnehmosclaudeassisted
3mo ago+100 -2
47%Typical
more lines added than baseline
22
d93600d8docs: update README for v0.2 architecture
Mnehmosclaudeassisted
3mo ago+68 -23
42%Typical
more lines added than baseline
23
ee4660e9refactor(app): wire v0.2 pipeline into main loop, drop WebGazer
Mnehmosclaudeassisted
3mo ago+66 -36
42%Typical
more lines added than baseline
24
081b8605feat(calibration): scanning-dot collection + speed trims
Mnehmosclaudeassisted
3mo ago+63 -11
38%Typical
more lines added than baseline
25
9b1847f3feat(theme): cream-white lighting mode — monitor as diffused key light
Mnehmosclaudeassisted
3mo ago+75 -7
36%Typical
more lines added than baseline
26
278fa083fix(calibration): dot-placement timing + persistent webcam preview
Mnehmosclaudeassisted
3mo ago+68 -24
33%Below typical
more lines added than baseline
27
0052b8affeat(config): add v0.2 tunables for MediaPipe, classifier, calibration
Mnehmosclaudeassisted
3mo ago+52 -15
30%Below typical
more lines added than baseline
28
e8fc7a04feat(gaze): EMA smoothing + denser grid for side-eye accuracy
Mnehmosclaudeassisted
3mo ago+44 -10
26%Below typical
more lines added than baseline
29
423f9497feat(calibration): smooth bell-envelope dot animation
Mnehmosclaudeassisted
3mo ago+45 -59
26%Below typical
more lines added than baseline
30
b1e617bcfeat(calibration): stationary dots, denser grid, realistic threshold
Mnehmosclaudeassisted
3mo ago+32 -68
21%Below typical
more lines added than baseline
31
dcf1d799chore(v0.1): commit pending favicon + MediaPipe CDN patch before v0.2
Mnehmosclaudeassisted
3mo ago+59 -6
21%Below typical
more lines added than baseline
32
61a511b3feat(telemetry): surface classifier confidence and head pose
Mnehmosclaudeassisted
3mo ago+36 -0
17%Below typical
more lines added than baseline
33
506ad8cbfeat(export): include classifier + normalization metadata in v0.2 export
Mnehmosclaudeassisted
3mo ago+13 -3
14%Below typical
more lines added than baseline
34
b43d2d33fix(deps): bump @mediapipe/tasks-vision to 0.10.34
Mnehmosclaudeassisted
3mo ago+2 -2
12%Below typical
fewer lines added than baseline
35
95b22a4bfix(calibration): second stale accuracy reference in return statement
Mnehmosclaudeassisted
3mo ago+1 -1
7%Below typical
fewer lines added than baseline
36
2cc158a9fix(calibration): validation threshold check used the old var name
Mnehmosclaudeassisted
3mo ago+1 -1
7%Below typical
fewer lines added than baseline
37
b8d4daedfix(calibration): rename leftover contentBricks → contentBricksAll
Mnehmosclaudeassisted
3mo ago+1 -1
7%Below typical
fewer lines added than baseline
38
f436cae6chore(css): remove dead WebGazer-specific selectors from app.css
Mnehmosclaudeassisted
3mo ago+0 -19
1%Below typical
fewer lines added than baseline
  • 5c4d932ffeat(beta): initial v0.1 release - gaze-tracking pedagogy instrument with brick content model, typed event detection, live telemetry, and two starter lessons (gravity, recursion)
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +2.8K -0
    Risk
    99%Elevated
  • 6a089b35feat(ensemble): multi-head classifier — train multiple versions at once
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +542 -389
    Risk
    95%Elevated
  • 6bd0c6c0feat(gaze-mode): add regression mode — predict viewport coords, hit-test
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +420 -210
    Risk
    95%Elevated
  • be4ea95dPivot gaze pipeline to classification-first
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +355 -152
    Risk
    89%Elevated
  • 8a71a066feat(calibration): replace 9-point with per-brick classifier calibration
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +420 -91
    Risk
    89%Elevated
  • c7a53dc6test(fixtures): synthetic landmark generator + DOM test harness
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +273 -0
    Risk
    84%Elevated
  • f01ec944feat(features): geometric feature extraction for gaze classifier
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +297 -0
    Risk
    84%Elevated
  • 1e9fd92ffeat(reader-cursor): mouse-driven reading + word-level tracking
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +347 -5
    Risk
    80%Elevated
  • 2165909efeat(classifier): TensorFlow.js MLP for feature→brick classification
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +226 -0
    Risk
    78%Elevated
  • 8c5ef2f9feat(calibration): grid method — viewport-fixed dots, no scroll
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +193 -9
    Risk
    72%Elevated
  • 6e9f2be3docs: update PROJECT_KNOWLEDGE.md for v0.2
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +189 -46
    Risk
    72%Elevated
  • 9cf74c79feat(face-landmarker): add MediaPipe Face Landmarker wrapper
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +189 -0
    Risk
    72%Elevated
  • 0bad03e5feat(export): LLM-grade reading trace — word-level data + event sourcing
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +160 -27
    Risk
    64%Typical
  • 2172a91dfeat(pagination): page field on bricks, no auto-scroll, tighter classifier gates
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +234 -60
    Risk
    64%Typical
  • f8ca2e9adocs(handoff): add VALIDATION_HANDOFF.md for Vario
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +182 -0
    Risk
    64%Typical
  • 7072dd8dfeat(gaze): rewrite pipeline to consume classifier output
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +140 -35
    Risk
    59%Typical
  • 9182c1e9test(pipeline): integration test for Features → Classifier → Gaze
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +135 -0
    Risk
    57%Typical
  • f405966ffeat(events): dwell-gated first_read + mouse-last-word + gaze override
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +108 -10
    Risk
    54%Typical
  • d3341778test(features): unit tests for feature extraction + normalization
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +115 -0
    Risk
    51%Typical
  • c8a90347test(classifier): unit tests for MLP build/train/predict/argmax
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +97 -0
    Risk
    47%Typical
  • 7e5f6fdcfeat(deps): remove WebGazer, add TensorFlow.js CDN + v0.2 module slots
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +100 -2
    Risk
    47%Typical
  • d93600d8docs: update README for v0.2 architecture
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +68 -23
    Risk
    42%Typical
  • ee4660e9refactor(app): wire v0.2 pipeline into main loop, drop WebGazer
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +66 -36
    Risk
    42%Typical
  • 081b8605feat(calibration): scanning-dot collection + speed trims
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +63 -11
    Risk
    38%Typical
  • 9b1847f3feat(theme): cream-white lighting mode — monitor as diffused key light
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +75 -7
    Risk
    36%Typical
  • 278fa083fix(calibration): dot-placement timing + persistent webcam preview
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +68 -24
    Risk
    33%Below typical
  • 0052b8affeat(config): add v0.2 tunables for MediaPipe, classifier, calibration
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +52 -15
    Risk
    30%Below typical
  • e8fc7a04feat(gaze): EMA smoothing + denser grid for side-eye accuracy
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +44 -10
    Risk
    26%Below typical
  • 423f9497feat(calibration): smooth bell-envelope dot animation
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +45 -59
    Risk
    26%Below typical
  • b1e617bcfeat(calibration): stationary dots, denser grid, realistic threshold
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +32 -68
    Risk
    21%Below typical
  • dcf1d799chore(v0.1): commit pending favicon + MediaPipe CDN patch before v0.2
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +59 -6
    Risk
    21%Below typical
  • 61a511b3feat(telemetry): surface classifier confidence and head pose
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +36 -0
    Risk
    17%Below typical
  • 506ad8cbfeat(export): include classifier + normalization metadata in v0.2 export
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +13 -3
    Risk
    14%Below typical
  • b43d2d33fix(deps): bump @mediapipe/tasks-vision to 0.10.34
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +2 -2
    Risk
    12%Below typical
  • 95b22a4bfix(calibration): second stale accuracy reference in return statement
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +1 -1
    Risk
    7%Below typical
  • 2cc158a9fix(calibration): validation threshold check used the old var name
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +1 -1
    Risk
    7%Below typical
  • b8d4daedfix(calibration): rename leftover contentBricks → contentBricksAll
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +1 -1
    Risk
    7%Below typical
  • f436cae6chore(css): remove dead WebGazer-specific selectors from app.css
    Author
    Mnehmos
    When
    3mo ago
    Lines
    +0 -19
    Risk
    1%Below typical
Showing 38 of 38 commits

How the score behaves here

Change risk →

Two views of the same model: where the cuts fall, and what commit shape lands you above them.

Score distribution

Every scored commit, binned on the raw 0 to 10 score rather than the percentile. Percentile ranks are uniform by construction, so that axis has no shape to draw. The dashed lines are the tercile cuts behind each row's priority pill.

08typical ↑↑ elevated0.05.010.0Change-risk score →
Below typical
Typical
Elevated

Size against diffusion

The 38 most recent commits, on their own recency sample rather than the feed above: that defaults to risk-sorted, so reusing it would plot only the top tercile and call it the spread. Big and scattered is what the model penalises. Click a dot to open it.

1101001,000Lines changed (log) →0.03.9Diffusion →
Below typical13
Typical13
Elevated12

Commit history for Mnehmos/mnehmos.oculus.webapp

Mnehmos/mnehmos.oculus.webapp has 38 commits in its history from 1 contributor, the first of them Apr 19, 2026. No file in the repository changed in the last 90 days. Every commit is scored for change risk from its size, spread and the history of the files it touches.