01 · Roasts
Sprint God, Rest God
38 straight weeks of zero commits, then 14 weeks of daily 4-intensity bursts. Your GitHub heatmap looks like someone forgot to plug in the keyboard until March 2026.
Repo Blitz Speedrun
port-harbour: 7 days. zagrosi-forge: 2 days. vrp-hunt: 2 days. kestrel: 8 days. You're shipping full-stack projects faster than most people write the README — which raises the question: are these shipped, or just speed-run?
The Profile Repo
zagrosi-code: 3KB, 0 stars, no source code, no license, no tests, no CI. You built a whole Rust project management platform with OIDC/SAML and fuzz harnesses, but your own profile page is an empty SVG folder.
ARCHITECTURE.md Collector
Five repos each ship ARCHITECTURE.md, DESIGN.md, STATUS.md, and SECURITY.md. The docs are longer than some of the codebases. At some point the architecture doc IS the project.
Community of One
6 followers, 0 forks across all 20 repos, and 11 PRs to other projects. You're building in a bunker — impressive output, but GitHub thinks you're a ghost.
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% weight56D
- Consistency20% weight60C
- Quality20% weight72B
- Depth15% weight55D
- Breadth10% weight72B
- Community10% weight28F
03 · Stats
365-day commit heatmap
75 active days
Language distribution
- Python36%
- Rust33%
- TypeScript10%
- Java7%
- HTML5%
- CSS4%
- Other5%
04 · Numbers
Owned repos
non-fork
18
Commits
last 12 months
152
Followers
6
Joined GitHub
Mar 2022
05 · Top repos
zagrosi-code /
port-harbour
Polished VS Code extension + Rust TUI tool visualizing system ports as pixel-art harbour. Dual-platform (Node+React webview, Rust CLI), typed, documented, with CI but nascent (7 days old, 4 stars). Shipped to marketplace with coherent UX.
zagrosi-code /
zagrosi-forge
Zagrosi Forge is a Codex-native plugin providing project decomposition, TDD planning, and implementation workflows. It ships with comprehensive Python CLI helpers, tests, CI/CD, structured documentation, and example SaaS case studies demonstrating the full workflow end-to-end.
zagrosi-code /
zagrosi
Pre-alpha self-hosted project management platform in Rust. Early-stage with solid architecture foundation (identity layer, cryptography, OIDC/SAML, error handling), typed codebase, comprehensive CI, and good documentation. 23 commits in 10 days, but very young and not yet usable; limited external adoption.
zagrosi-code /
kestrel
Educational research project on browser automation and anti-bot evasion. Typed Python (FastAPI/Pydantic), comprehensive docs (DESIGN.md, ARCHITECTURE.md, PRIVACY.md, SECURITY.md), ~200 LOC mock site, 30+ commits across 8 days, rigorous conventions; experimental scope and tiny stars.
zagrosi-code /
vrp-hunt
Specialized Python toolkit for Google VRP research with guardrails, recon modeling, and autonomous agent planning. Typed, well-documented, and tested but newly published (2 days old) with minimal adoption signals.
zagrosi-code /
Starling-Bank-Round-Up
Java fintech API integration project implementing Starling Bank round-up savings feature. Typed, documented, modular design with service architecture; lacks tests, CI, production exposure, and has no license. Appears to be a coding challenge response.
zagrosi-code /
capslock-nodelay
Shell script utility that removes macOS Caps Lock delay via LaunchAgent. Very narrow scope, minimal code, 3 stars, recent burst of 3 commits in 4 days. No tests or CI. Clear single-purpose tool with MIT license and working install/uninstall.
zagrosi-code /
zagrosi-code
Minimal portfolio/resume repo with only README and assets; no source code, tests, CI, or license. Purely presentational profile page with no functional depth or architectural scope.
06 · Timeline
- Mar 15, 2022Joined GitHub
- Mar 5, 2025Created Starling-Bank-Round-Up
- May 7, 2026Created zagrosi-code — Profile
- May 7, 2026Created zagrosi — Open-source, self-hosted, AI-native platform for project management, issue tracking, and incident response.
- May 10, 2026Created kestrel — Educational research project: human-in-the-loop browser automation for personal-use UK car insurance quote shopping.
- May 12, 2026Created port-harbour — Live pixel-art harbour for your dev ports + Docker containers. Ports = boats. VS Code extension + Rust TUI.
- May 13, 2026Created capslock-nodelay — Remove the macOS Caps Lock activation delay via a LaunchAgent that runs hidutil at login.
- May 16, 2026Created vrp-hunt — Guardrailed tooling for authorized VRP research workflows
- May 17, 2026Created zagrosi-forge — Codex-native project decomposition, planning, quality gates, traceability, and TDD implementation skills.
- May 19, 2026Most recent push to port-harbour
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.