▸ This tool was built by an AI agent from Zoral
← RATE MY GITHUB

#72 — Top 94.0%

CypherGuy

Kabir Ghai

B

Solid engineer

Overall

0.0

/ 100

01 · Roasts

The Zero-Star Collector

7 named projects, 0 stars on 8 of 9 repos. You're building impressively complex systems — WAL-based KV stores, multi-agent AI pipelines, JWT-secured full-stack apps — entirely for an audience of one. URL-Shortener's 14 stars are carrying the entire portfolio's social proof.

Burst Mode Developer

ADK-Agent went from zero to ARCHITECTURE.md + 90 tests in 6 days. Ripple was 30 commits in 10 days for a hackathon. Your heatmap looks like an EKG — long flatlines interrupted by cardiac events. Sustained daily grind: not your style.

Solo Artist

soloPct = 98%. Five PRs in a year. You've built a full-stack health app with S3 integration, a systems-level KV store, and a multi-agent hackathon entry — all completely alone. GitHub is a collaboration platform you're using as a personal diary.

Type System Amnesia

You typed BloodReview, QueryWhisperer, StudyBattles — all great. Then kv-engine, ADK-Agent, PhaseCompiler, and procedural-generation all ship as raw untyped Python. The pattern suggests type annotations are a mood, not a discipline.

47% Graveyard Rate

staleRepoRatio = 0.47. Nearly half your 48 repos haven't been touched in 2+ years. That's a lot of README.md files quietly aging in the dark, whispering 'coming soon' to nobody.

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

  • Impact
    25% weight
    62C
  • Consistency
    20% weight
    65C
  • Quality
    20% weight
    72B
  • Depth
    15% weight
    65C
  • Breadth
    10% weight
    65C
  • Community
    10% weight
    55D

03 · Stats

365-day commit heatmap

125 active days

Less
More

Language distribution

6 langs
  • Python53%
  • TypeScript21%
  • JavaScript13%
  • HTML9%
  • CSS3%
  • PowerShell1%

04 · Numbers

Owned repos

non-fork

43

Commits

last 12 months

593

Followers

12

Joined GitHub

Nov 2020

05 · Top repos

CypherGuy /

URL-Shortener

60/100

Production-grade URL shortener educational project with multi-section architecture progression, comprehensive test suite, CI/CD, AWS deployment with load-tested infrastructure (2000 concurrent users at 0.008% failure rate). Typed Python with structured layout, extensive docs including AWS_ARCHITECTURE.md and design spe

I40Q75D65
READMETestsCI
Python020d ago

CypherGuy /

Ripple

55/100

Google Cloud hackathon submission: multi-agent AI system on FastAPI + Google ADK that correlates PR diffs with Dynatrace incidents to auto-generate fix MRs. 30 commits over 10 days; typed Python, structured 4-service architecture, comprehensive tests, live dashboard with WebSocket broadcasting.

I55Q60D50
READMETests
Python09d ago

CypherGuy /

QueryWhisperer

50/100

Full-stack AI-to-SQL web app with typed FastAPI backend (Python), Next.js frontend, PostgreSQL integration, JWT auth, and CI/CD. Modular architecture with meaningful documentation and early-stage development trajectory.

I40Q60D50
READMECITyped
Python120d ago

CypherGuy /

StudyBattles

50/100

Gamified study tool with FastAPI backend + React frontend. Full-stack typed project with tests and CI, structured multi-file layout, clear architecture. Early-stage but working implementation of document parsing, LLM-driven question generation, and unlock mechanics.

I40Q60D50
READMETestsCITyped
Python020d ago

CypherGuy /

BloodReview

50/100

Functional full-stack health app with blood test tracking, CSV import, charting, and Claude AI analysis. Typed Node.js/Express backend, secure auth (bcrypt, CSP headers), MySQL integration, S3 upload, CI/test coverage, and clear README with setup docs.

I40Q60D50
READMETestsCITyped
HTML020d ago

CypherGuy /

procedural-generation-practical

50/100

Educational procedural generation project with deterministic tile-based map generation, comprehensive testing, and CI. Implements custom LCG PRNG with BFS validation and probabilistic lake/river formation using pygame.

I25Q60D50
READMETestsCI
Python020d ago

CypherGuy /

kv-engine

50/100

Educational single-process KV store with crash-safe persistence using WAL, atomic snapshots, checkpointing, and linearizable concurrency. Well-documented systems project exploring durability and correctness from first principles, but untyped Python with no CI/tests runner integration.

I25Q60D65
READMETests
Python020d ago

CypherGuy /

ADK-Agent

47/100

A specialized Python CLI using Google ADK agents to surface project ideas by cross-referencing underused technologies with newsletter trends. Well-documented with tests and CI, but early-stage with no external adoption or stars.

I25Q65D50
READMETestsCI
Python020d ago

CypherGuy /

PhaseCompiler

42/100

A Claude skill for generating phased project roadmaps with structured tasks and commit conditions. Typed Python project with meaningful documentation (SKILL.md, schema.py) and self-improvement loop (autoeval/). Limited stars/adoption but demonstrates serious craftmanship in prompt engineering and evaluation infrastruct

I25Q50D45
READMECI
Python220d ago

06 · Timeline

  1. Nov 5, 2020
    Joined GitHub
  2. Mar 30, 2024
    Created BloodReview — A website used to track, analyse and evaluate Blood Test results
  3. Feb 28, 2025
    Created QueryWhisperer — AI-powered web app that converts natural language into SQL using schema-aware, modular backend design.
  4. Dec 13, 2025
    Created kv-engine — A durable, concurrent key–value storage engine using write-ahead logging, atomic snapshots, and deterministic crash recovery.
  5. Feb 24, 2026
    Created PhaseCompiler — An agent skill for generaing roadmaps for coding projects by splitting it up into phases
  6. Feb 28, 2026
    Created StudyBattles
  7. Mar 8, 2026
    Created procedural-generation-practical — A mini project to help me understand the fundamentals of procedurally generated 'worlds'
  8. Apr 12, 2026
    Created URL-Shortener — A URL shortener built in stages. Each section adds one architectural layer with load test results to validate the tradeoff
  9. May 8, 2026
    Created ADK-Agent — A CLI tool using Google's ADK to provide project ideas based on repos and newsletters
  10. May 15, 2026
    Created Ripple — My submission for Google Cloud Rapid Agent Hackathon 2026
  11. May 25, 2026
    Most recent push to Ripple

07 · Compare

github.com/
CypherGuy · 6dmedian coder

08 · Rubric

How this score was produced

Overall = Σ (category × weight) + gentle top-end curve

CategoryWeightScoreContrib.
Raw total64.7
Top-end curve+5.6
Final overall70.3

Tier thresholds

S90100Mass-producing humansA8089Ship machineB7079Solid engineerC6069Getting thereD4059README enthusiastF039GitHub tourist
▸ How the pipeline works
  1. 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.
  2. 02Triage.A small model reads every repo's file tree + README and picks the 20 files per repo that actually reveal how you code.
  3. 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.
  4. 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.
  5. 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.
CypherGuy · 70.3/100 — Rate My GitHub