01 · Roasts
2,764 Commits and Still Shipping Stubs
You pushed 2,764 commits this year and somehow still found time to create blessed-car-apps (0 files) and demo-qr-generator (abandoned in 31 minutes). The heatmap is a wall of green but some of those squares are suspiciously hollow.
1,523 PRs/Year Is Not a Human Number
You opened 1,523 pull requests in a single year. That's 4.2 PRs per day, every day, including weekends. Either you've automated your own GitHub activity or you are the objective function at scalingforever.com.
License? Never Heard of Her
codex-autorunner has 746 stars and moes-tavern has a full ERC-8004 smart contract suite — but multiple repos are shipping without a LICENSE file. You're out here letting the world use your agent orchestration stack under the legal framework of 'vibes.'
The 4-Following Ratio
183 followers, following 4 people. You have a follower-to-following ratio of 45:1. Either you're a GitHub celebrity or you've fully optimized away social reciprocity in pursuit of the objective function.
Agent Orchestration All the Way Down
codex-autorunner supervises agents, agent-nexus coordinates agents, social-autorunner compiles agent campaigns, moes-tavern is a marketplace for agents, and memc is a memory compiler for agents. David, are you building tools or are the tools building you?
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% weight76B
- Consistency20% weight95S
- Quality20% weight72B
- Depth15% weight70B
- Breadth10% weight80A
- Community10% weight65C
03 · Stats
365-day commit heatmap
323 active days
Language distribution
- Python46%
- C27%
- TypeScript8%
- Go7%
- JavaScript6%
- Svelte2%
- Other4%
04 · Numbers
Owned repos
non-fork
38
Commits
last 12 months
2,764
Followers
183
Joined GitHub
Oct 2014
05 · Top repos
Git-on-my-level /
codex-autorunner
Production-grade Python agent coordination hub with 746 stars, full type safety, comprehensive test/CI suite, multi-surface architecture (web, CLI, chat), and extensive design documentation enabling long-running autonomous workflows across Codex/Hermes/OpenCode.
Git-on-my-level /
agent-nexus
Well-architected agent coordination monorepo with typed Go backend, SvelteKit UI, and Python bridge runtime. Shipped self-hosted workspace product with comprehensive infrastructure (tests, CI, docs, contracts). 30+ commits over ~2 months shows sustained multi-service development.
Git-on-my-level /
ios-mosh-terminal
iOS Mosh terminal client implementing SSH bootstrap + UDP-based Mosh with managed tmux and auto-reconnect. Typed Swift, structured services, binding design docs (CONSTITUTION.md), CI/tests, ~47 days old, 1 star. Non-trivial portfolio project with shipped architecture and working code.
Git-on-my-level /
moes-tavern
ERC-8004 agent marketplace with typed Solidity contracts + TS packages (curation, search, indexer). Full monorepo with tests, CI, but nascent (created Feb 2026, 18 commits in 1 day, 0 stars).
Git-on-my-level /
demo-web-agent-harness
A self-contained web agent harness demo in vanilla JS with a sandboxed LLM tool-execution loop. Typed config, modular architecture, and persistence—but no tests, CI, or production signals. Personal experimental project showing working proof-of-concept.
Git-on-my-level /
memc
Early-stage personal project: memory compiler for agent workflows. Typed Rust with structured docs (ARCHITECTURE.md, LANGUAGE_SPEC.md), CLI tool, parser, compiler, and recall engine—shipped as v0.1.0 with CI/release workflow. Non-trivial scope but nascent (0 stars, <30 commits).
Git-on-my-level /
crimson-web-port
TypeScript+Vite web port of Crimson game using Phaser 3, with documented architecture, CI/tests, and live demo—early-stage personal project showing structured development but nascent adoption.
Git-on-my-level /
social-autorunner
Python CLI tool for social media campaign management with structured compilation workflow, SQLite backend, LLM provider integration, and comprehensive config schema. Early-stage project (1 star, <2 months old) with working typed code, good test coverage, and published documentation.
Git-on-my-level /
vibe-evm
Python EVM investigation SDK designed for LLM agent workflows. Includes workspace CLI, typed clients (RPC/Etherscan/Tenderly), caching, budgeting, and artifact management. Well-documented with ARCHITECTURE.md and design patterns, but nascent (2 stars, minimal recent commits, no tests/CI).
Git-on-my-level /
car-ticket-templates
Minimal template repository for Codex Autorunner (CAR) with plain Markdown ticket templates. 2 stars, 6 commits in 2.5 months, no tests/CI/license. README explains purpose and conventions, but lacks code samples or deep architecture.
Git-on-my-level /
demo-qr-generator
Minimal HTML project with zero documentation, tests, or CI. Created and abandoned within 31 minutes with 3 commits. No meaningful architectural scope or ongoing investment detected.
Git-on-my-level /
blessed-car-apps
Empty repository with initial commit only, no files, no documentation, no code. Not yet a project.
06 · Timeline
- Oct 24, 2014Joined GitHub
- Dec 9, 2025Created codex-autorunner
- Dec 25, 2025Created vibe-evm — A prescriptive way for LLM agents to analyze EVM state and transactions
- Jan 27, 2026Created ios-mosh-terminal
- Feb 1, 2026Created car-ticket-templates
- Feb 4, 2026Created crimson-web-port
- Feb 6, 2026Created social-autorunner
- Feb 11, 2026Created moes-tavern — ERC-8004-aligned agent marketplace
- Mar 5, 2026Created agent-nexus
- Apr 17, 2026Created demo-qr-generator — Fun QR code generator — custom colors, sizes, WiFi presets, PNG/SVG export
- Apr 17, 2026Created demo-web-agent-harness — Self-modifying web agent demo, a self contained web agent harness
- Apr 21, 2026Created memc
- Apr 29, 2026Created blessed-car-apps
- Apr 30, 2026Most recent push to codex-autorunner
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.