01 · Roasts
README? Never Heard of Her
about-blank has 3.5 years of commits and 804 KB of code. Its README says '# start'. That's it. Three and a half years and you wrote two words. One of them is a CSS value.
229 PRs, 99% Solo
You opened 229 PRs this year but soloPct is 99%. Those are self-PRs. You're having pull request conversations with yourself at industrial scale. Therapy is also a valid workflow.
Perpetual Early Access
inference-provider-registry, browser-terminal, luna, proxy-up — all fresh, all ambitious, all 0 stars. You're not shipping products, you're shipping product announcements to an empty room.
391 Stars, One Repo's Doing 51% of the Work
aghub has 200 of your 391 total stars. Strip that out and your entire public portfolio — 290+ repos — averages 0.65 stars each. The long tail is very, very long.
3,266 Commits and Still No Tests in 9 of 12 Repos
You commit more in a week than most developers do in a year, yet 75% of your sampled repos have HAS_TESTS=no. Quantity of commits is not a substitute for confidence that the code works.
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% weight68C
- Consistency20% weight95S
- Quality20% weight69C
- Depth15% weight65C
- Breadth10% weight72B
- Community10% weight65C
03 · Stats
365-day commit heatmap
361 active days
Language distribution
- TypeScript56%
- Rust38%
- JavaScript3%
- CSS1%
- Less0%
- Go0%
- Other2%
04 · Numbers
Owned repos
non-fork
94
Commits
last 12 months
3,266
Followers
321
Joined GitHub
Jun 2021
05 · Top repos
AkaraChen /
aghub
Unified MCP/skill configuration hub for 22+ AI coding agents. Typed Rust (200 stars), documented (README + docs/ + ARCHITECTURE.md), structured multi-crate layout (agents, core, cli, api, desktop), CI/CD active. No unit tests HAS_TESTS=no, but shipping real product with Tauri desktop app and REST API.
AkaraChen /
nashi
Modern DOM manipulation library with jQuery-like chaining API. TypeScript-typed, documented, tested, and published to npm. Modest adoption (63 stars) with clean architecture, though no CI/CD automation present.
AkaraChen /
2code
Tauri + React + TypeScript desktop workstation with integrated terminal, git workflows, and file browsing. Typed, tested, CI-enabled, well-documented with design docs and proper multi-tier architecture across frontend and Rust backend.
AkaraChen /
proxy-up
Typed, well-documented gateway orchestration tool for LLM proxying (Go + TypeScript). Structured monorepo with tests, config system, and comprehensive docs (design.md, ARCHITECTURE.md). 494KB, ~30 recent commits over 1 month. No CI/license but working production-ready foundation.
AkaraChen /
browser-terminal
Early-stage local browser terminal in Rust using Axum, WebSocket, and Xterm.js. Well-documented with typed code and security-conscious architecture, but extremely fresh (1 day old) with minimal adoption signals.
AkaraChen /
luna
A personal Rust daemon project integrating GitHub Projects with coding agents. Pre-production (heavy construction), minimal adoption (1 star), but demonstrates architectural ambition: typed codebase, multi-crate Cargo layout, documented workflow orchestration.
AkaraChen /
inference-provider-registry
Typed TypeScript registry with CI pipeline for AI model providers. Has README, schema validation, and automated daily model updates via GitHub Actions, but zero adoption signals and minimal commit history over short window.
AkaraChen /
hazuki
Vue 3 + Vite template with 1GB codebase, 1+ year active commits, but minimal docs (generic scaffold README), no tests/CI/license, and 2 stars indicate experimental personal project.
AkaraChen /
next-progress
TypeScript React hook wrapper around NProgress for Next.js route progress indication. Minimal utility with proper types, config, and documentation but no tests and minimal commits (3 of last 30).
AkaraChen /
about-blank
Personal TypeScript project with minimal documentation (README contains only "# start"), no tests or CI, 804 KB codebase with 30 recent commits, but no clear output or functional scope visible.
AkaraChen /
proxy-in-menubar
Rust Tauri macOS menubar app example with MIT license. No tests, no CI, minimal commit activity (3 of last 30), but structured codebase (4250 KB suggests 80–120 LOC) and README present.
AkaraChen /
react-components
Minimal TypeScript React component library with 0 stars, no tests or CI, thin documentation (README only), and 74 KB codebase. Published to npm as @akrc/rc but shows no adoption or community engagement.
06 · Timeline
- Jun 1, 2021Joined GitHub
- Oct 2, 2022Created about-blank
- Nov 1, 2022Created nashi — 🍐 The modern jQuery alternative.
- Mar 19, 2023Created react-components
- Mar 25, 2023Created next-progress
- May 25, 2025Created hazuki
- Feb 10, 2026Created 2code — The Vibe Coding Workstation — A desktop workspace where terminal, AI agents, and git live together for uninterrupted flow state
- Mar 2, 2026Created proxy-in-menubar
- Mar 16, 2026Created aghub — One hub for every AI coding agent. Unified configuration management for 22+ assistants.
- Apr 4, 2026Created proxy-up
- Apr 7, 2026Created inference-provider-registry
- Apr 29, 2026Created luna — Long-running coding-agent orchestrator for GitHub Projects
- May 4, 2026Created browser-terminal — A local browser-based terminal powered by Rust, PTY forwarding, WebSocket, and Xterm.js.
- May 6, 2026Most recent push to inference-provider-registry
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.