01 · Roasts
One Star to Rule Them All
384 of your 511 total stars live in bulletty. Strip that one Rust TUI app and your entire portfolio averages 1.3 stars per repo. You're not a multi-product shipper — you're a one-hit wonder with 69 side quests.
The Eternal Non-Tester
Zero test files across all 11 sampled repos. Not one. bulletty ships on Homebrew, Arch, and Snap with CI badges and still no tests. You're out here distributing to package managers on pure vibes.
README? More Like READ-MAYBE
crocidb-clutterlog's entire README is 'This is my clutterlog.' feeds has no README at all. You built a blog you've maintained for 10 years and somehow couldn't find 3 sentences to describe a sister project.
Graveyard Curator
65% of your 70 repos haven't been touched in over 2 years. That's 45+ abandoned projects. At some point 'tinkerer' is just a polite word for 'someone who names repos and moves on.'
Vibe-Coded and Proud
obsidian_to_blog.py is self-described as 'vibe-coded', uses regex-based HTML parsing, and was created and abandoned on the same day in under 3 minutes of commits. The README claims MIT license but HAS_LICENSE=no. The vibes did not include a LICENSE file.
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% weight63C
- Consistency20% weight65C
- Quality20% weight69C
- Depth15% weight65C
- Breadth10% weight80A
- Community10% weight55D
03 · Stats
365-day commit heatmap
252 active days
Language distribution
- C#42%
- C14%
- ShaderLab12%
- C++9%
- Assembly5%
- JavaScript4%
- Other14%
04 · Numbers
Owned repos
non-fork
54
Commits
last 12 months
698
Followers
140
Joined GitHub
Apr 2009
05 · Top repos
crocidb /
bulletty
Bulletty is a mature Rust TUI RSS/Atom feed reader with 384 stars, shipped across multiple package managers (Arch, Homebrew, Snap, Nix), strong project documentation, CI/CD, and multi-module architecture. Lacks automated tests despite active development.
crocidb /
clutterlog
Early-stage Rust static site generator for media galleries with clean architecture, comprehensive README, CI setup, and typed code. No stars yet; created Feb 22, 2026, with ~30 commits showing focused development but limited external adoption.
crocidb /
annileen
Early-stage 3D game engine in C++ using bgfx, with asset pipeline and voxel support. Typed language, CI (Travis), README, structured multi-file layout (~62MB codebase), but no unit tests and limited adoption signals.
crocidb /
blog
Personal blog built with Hugo in Nix, with 10 years of continuous development (2016–2026) and a live domain (crocidb.com), but minimal discoverability (0 stars/forks) and no typed code or tests.
crocidb /
funbox-experiments
Personal audio effects experiments for a custom FunBox guitar pedal, with FM synthesis implementation in C++. Well-documented build/dev setup, but nascent scope (0 stars, 10 commits in 3 months, no tests/CI).
crocidb /
dotfiles
Personal dotfiles repo with Neovim, Emacs, and Tmux configs using Lua and Elisp. Well-structured setup with plugin configurations but minimal documentation and no tests/CI.
crocidb /
remote-tactical-wars
Game jam submission built with Three.js. Untyped JavaScript, structured multi-file layout, documented via README, but no tests, CI, or license. 30 commits over 4 days suggest concentrated effort on a complete puzzle game with 8 levels.
crocidb /
crocidb-clutterlog
Personal clutterlog project with minimal documentation and no clear purpose. ~200KB codebase with 18 commits in ~18 days, but lacks tests, license, types, and substantive README describing the project.
crocidb /
obsidian-to-blog
Personal one-off tool for converting Obsidian notes to Hugo blog posts. Single Python file (~150 LOC), minimal dependencies, working but unpolished. Created and committed within 3 minutes (2 commits in last 30 days).
crocidb /
wayland-timelapse
Single shell script wrapper for Wayland screencasting; minimal scope (2KB), one commit, no tests/CI/license. Functional guide but nascent project with no architectural depth.
crocidb /
feeds
Shell-based personal feeds project with minimal documentation, no tests, CI, or structure. 30 commits over ~6 months but unclear purpose and no public README guidance.
06 · Timeline
- Apr 5, 2009Joined GitHub
- Mar 28, 2013Created dotfiles — my dotfiles
- Oct 18, 2016Created blog — My blog where I talk about subjects of my interest, such as programming, music, languages, photography, etc.
- Apr 9, 2018Created annileen — a toy 3d game engine with voxel support
- May 29, 2025Created bulletty — bulletty is a pretty feed reader for the terminal that stores the articles as Markdown
- Nov 1, 2025Created feeds — my feeds with bulletty
- Nov 16, 2025Created funbox-experiments — some audio effects that I'm building for my custom FunBox pedal
- Feb 22, 2026Created clutterlog — static gallery site generator focused on sharing your projects's WIPs and creative process
- Feb 22, 2026Created crocidb-clutterlog
- Mar 12, 2026Created obsidian-to-blog — Convert Obsidian draft note to a Hugo blog post
- Apr 16, 2026Created wayland-timelapse — create timelapses from your screens in wayland
- Apr 18, 2026Created remote-tactical-wars — Game made for the LudumDare 59
- Apr 27, 2026Most recent push to dotfiles
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.