01 · Roasts
README union shop
Every scored repo has a README, but several are title-only—documentation showed up; detail took the day off.
CI witness protection
Across 12 scored repositories, tests and CI are absent everywhere. The build pipeline remains an unsolved side quest.
Hackathon afterburner
CodeCraft logged 28 of 30 sampled commits, but its entire observed life is 2026-09-17 to 2026-09-18: impressive sprint, tiny runway.
Paws carries
GROUP-R-MULTIMEDIA is the portfolio heavyweight at 61,509 KB; the rescue site is hauling more than its share of the depth score.
Built using
Zoral
Shadows one worker for a week, then takes over their job with zero extra setup. Behaves exactly like the original.
zoral.ai
02 · Category breakdown
- Impact25% weight48D
- Consistency20% weight60C
- Quality20% weight28F
- Depth15% weight55D
- Breadth10% weight55D
- Community10% weight40D
03 · Stats
365-day commit heatmap
25 active days
Language distribution
- JavaScript56%
- HTML23%
- CSS15%
- Python4%
- C++1%
- C#1%
04 · Numbers
Owned repos
non-fork
13
Commits
last 12 months
137
Followers
12
Joined GitHub
Mar 2024
05 · Top repos
NjabuloNhlabathi /
GROUP-R-MULTIMEDIA
A substantial static PawsNelspruit animal-rescue site with six HTML pages, adoption and shop flows, 12 animal listings, and responsive CSS, but no backend behavior, tests, CI, license, or meaningful project documentation.
NjabuloNhlabathi /
UNITY-GAME
A small, documented Unity 2D learning game with C# player physics, randomized obstacles, score UI, and an Input System asset; it is an early one-day practice project without tests, CI, or external adoption.
NjabuloNhlabathi /
CodeCraft
Hackathon IoT prototype combining Arduino hazard control, a Flask serial sensor dashboard, and OpenCV-based underwater contamination scanning, but with minimal documentation and no visible validation or release infrastructure.
NjabuloNhlabathi /
Hand-tracker-python
A small, documented Python/OpenCV demo implementing real-time single-hand MediaPipe tracking, landmark drawing, and an index-finger-triggered image overlay in one script.
NjabuloNhlabathi /
UMP-food-link
A functional hackathon food-waste web app with separate student, staff, and homepage flows, local browser data handling, CSV-backed waste insights, and a polished neumorphic UI, but no tests, CI, license, or adoption evidence.
NjabuloNhlabathi /
Java-calculator
A one-day HTML prototype with a README, but the advertised Java calculator is not implemented: index.html presents an unrelated alert network, cafedashboard.html is nearly empty, and style.css has no styles.
NjabuloNhlabathi /
Njabulo-Group-R-alone
Small static animal-rescue landing page with a responsive HTML/CSS layout, slideshow, navigation, and footer contact details, but no demonstrated adoption, testing, or sustained repository history.
NjabuloNhlabathi /
NjabuloNhlabathi
A 98 KB, zero-star profile-style repository centered on a README tech-stack presentation; no source files, tests, CI, license, or concrete shipped product are evidenced.
NjabuloNhlabathi /
Adruino-LED-I-LOVE-YOU
A tiny Arduino LCD greeting sketch that renders a custom heart and “I LIKE YOU”/“SIPHELELE” message, with minimal project structure and no adoption signals.
NjabuloNhlabathi /
FocusFox
FocusFox is a one-commit Robo Mania hackathon scaffold with a README title and description but no fetched source files or implementation evidence.
NjabuloNhlabathi /
Air-Board
Air-Board is an empty initial scaffold: the README names a Python/MediaPipe/OpenCV concept, but no source files or implementation are present.
NjabuloNhlabathi /
Personal-Portfolio
Personal-Portfolio is an empty, one-commit scaffold: it contains only a minimal README and no source files, tests, CI, license, or product evidence.
06 · Timeline
- Mar 3, 2024Joined GitHub
- Jun 24, 2026Created UNITY-GAME — First practice game using unity
- Aug 13, 2026Created Java-calculator — java calculator
- Aug 14, 2026Created UMP-food-link
- Aug 24, 2026Created GROUP-R-MULTIMEDIA — GROUP R MULTIMEDIA
- Aug 24, 2026Created Personal-Portfolio
- Aug 28, 2026Created Njabulo-Group-R-alone — Njabulo personal group project
- Aug 29, 2026Created NjabuloNhlabathi
- Sep 5, 2026Created Adruino-LED-I-LOVE-YOU
- Sep 11, 2026Created Hand-tracker-python
- Sep 14, 2026Created Air-Board — build with python,media pipe and open cv
- Sep 16, 2026Created FocusFox — Robo mania hackathon project
- Sep 17, 2026Created CodeCraft — A IoT based solution for the MICT SETA Hackathon
- Sep 18, 2026Most recent push to CodeCraft
07 · Compare
08 · Rubric
How this score was produced
Overall = Σ (category × weight) + gentle top-end curve
Tier thresholds
▸ How the pipeline works
- 01Scrape.Pull every non-fork repo pushed in the last 90 days, plus your contribution calendar, followers, and language byte counts — straight from GitHub's REST & GraphQL APIs.
- 02Triage.A small model reads every repo's file tree + README and picks the 20 files per repo that actually reveal how you code.
- 03Grade each repo. All repos run in parallel through a fast scoring model that reads the picked files and rates each one independently on Impact, Quality, and Depth — with evidence citations.
- 04Aggregate. A larger reasoning model combines the per-repo scores with server-computed stats (heatmap, commit cadence, language entropy, follower count) to produce the 6-dimension profile score + roasts.
- 05Correct.Deterministic server-side checks enforce anchor-scale floors (e.g. a profile with 2,000+ public commits can't score 30 Consistency) and recompute the final verdict.
~90 seconds per profile, ~$0.25 in compute. Total of ~240 files read across your top-12 repos. One rating per GitHub account per day.
▸ Data sources & caveats
- Heatmap & commit totals: GitHub GraphQL
contributionsCollection— covers the last 365 days, includes private repos when the user has opted in (default). - Language %: byte totals across the top 30 owned non-fork repos.
- Curve: a small upward nudge centered on raw score ≈ 70, capping at 100. Prevents specialists from being unfairly penalised for narrow breadth.
- Anchor corrections: when server-measured signals (e.g. privateWorkLikely, multiRepoVolume, follower count) mandate a minimum category score, the aggregation step enforces it. These are signal-conditional, not identity-based floors.