01 · Roasts
following: 0
294 people follow you and you follow exactly zero back. Not mysterious — just the GitHub equivalent of reading everyone's diary without saying hello.
Burst-mode builder
Your heatmap is 40 weeks of tumbleweeds followed by a furious final sprint of max-density commits. Either you discovered caffeine in Q4 or you time-traveled from the future to ship 12 repos.
All repos, zero forks
gnhf has 1,673 stars and only 110 forks. People are starring your work like a museum exhibit — impressive, but nobody's taking it home to build on it.
Solo operator, maximum scope
soloPct=100 across 52 repos. You've built agent orchestrators, CLI frameworks, browser automation layers, and a grid combat game — all alone. Inspiring or concerning, depending on the on-call schedule.
Three-day unicorn factory
autopreso hit 232 stars in 3 days. gnhf was born in March and already has 1.6k. At this velocity you'll either IPO or burn out before the year is out.
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% weight65C
- Quality20% weight77B
- Depth15% weight72B
- Breadth10% weight65C
- Community10% weight55D
03 · Stats
365-day commit heatmap
99 active days
Language distribution
- Go62%
- TypeScript26%
- JavaScript9%
- CSS1%
- HTML1%
- GSC0%
- Other1%
04 · Numbers
Owned repos
non-fork
24
Commits
last 12 months
734
Followers
294
Joined GitHub
Jan 2013
05 · Top repos
kunchenguid /
gnhf
gnhf is a well-crafted agent orchestrator for autonomous coding loops with 1.6k stars, comprehensive TypeScript codebase, full test suite, and multi-agent support across Claude, Codex, Copilot, and ACP targets.
kunchenguid /
no-mistakes
Go CLI tool for AI-powered code review/testing/fixing via local git proxy. 496 stars, well-structured with typed code, comprehensive tests, CI, full docs site, and production-ready polish.
kunchenguid /
axi
TypeScript CLI design framework with 10 agent-ergonomic principles, benchmarked against MCP. 790 stars, well-documented with design.md/ARCHITECTURE.md, tests+CI, ~55KB codebase. Recent sustained development.
kunchenguid /
treehouse
Go CLI tool managing isolated git worktrees for AI agents. Well-architected with comprehensive documentation (design.md, ARCHITECTURE.md), strong testing (cmd/e2e_test.go), CI/CD pipeline, and cross-platform support (macOS/Linux/Windows), plus self-update mechanism.
kunchenguid /
ezoss
Go daemon for maintainer-side GitHub issue/PR triage automation with Claude/Codex agent integration, persistent SQLite state, and Bubble Tea TUI. Typed, tested, documented, multi-file architecture (~2MB codebase).
kunchenguid /
gh-axi
TypeScript CLI wrapper around GitHub's gh command designed for autonomous agents. Ships with tests, CI, types, and comprehensive documentation. 44 stars, ~30 recent commits across 1.5 months suggest modest indie project with clear shipped product.
kunchenguid /
autopreso
Indie speech-to-whiteboard tool with polished MVP: real-time Excalidraw canvas controlled by voice via LLM agent. Typed JS, comprehensive docs (AGENTS.md, design), CI+tests, npm package published. 232 stars, 3 days old with fresh commits.
kunchenguid /
chrome-devtools-axi
TypeScript CLI wrapper for browser automation (chrome-devtools-mcp) with TOON encoding, generation-tracked refs, and agent-ergonomic design. 75 stars, actively maintained, ships with tests, CI, types, and comprehensive documentation. Designed for AI agents.
kunchenguid /
trial-by-combat
Deterministic LLM 1v1 grid-based duel with HTTP API for agents. Well-structured game engine (src/engine.js ~1000 LOC), full test suite (6 test files), CI, TypeScript-less modern Node.js/ESM. Fresh launch (created 2026-04-28, 3 days old, 8 commits), 8 stars.
kunchenguid /
acp-mock
TypeScript testing utility for Agent Client Protocol that provides deterministic mock agent over stdio. Typed, well-tested, CI/CD enabled, published to npm at v1.1.0. Very recent project (May 2026) with minimal adoption but polished execution.
kunchenguid /
simplewords
Early-stage Chrome extension (AI-powered reply refinement) with solid TypeScript typing, comprehensive test coverage via vitest, and documented structure. Personal project shipping a working indie tool with release automation.
kunchenguid /
org-bench
TypeScript benchmarking tool with structured documentation (design.md, ARCHITECTURE.md, STATUS.md), tests, and CI. Young repo (8 days old) with sparse commit history (4 of 30) but solid architectural foundation despite missing README and license.
06 · Timeline
- Jan 10, 2013Joined GitHub
- Mar 14, 2026Created treehouse — Manage worktrees without managing worktrees.
- Mar 21, 2026Created axi — Design principles for agent ergonomics. Higher accuracy with lower token cost than both MCP and regular CLI.
- Mar 26, 2026Created gh-axi — GitHub CLI for agents — designed with AXI (Agent eXperience Interface).
- Mar 27, 2026Created chrome-devtools-axi — The most agent-ergonomic browser automation
- Mar 31, 2026Created gnhf — Before I go to bed, I tell my agents: good night, have fun
- Apr 5, 2026Created no-mistakes — git push no-mistakes
- Apr 16, 2026Created org-bench
- Apr 24, 2026Created ezoss
- Apr 28, 2026Created trial-by-combat — Two LLMs walk into a 9x9 grid. One walks out with the relic.
- May 2, 2026Created acp-mock — Test ACP clients E2E without spending tokens or trusting a live agent.
- May 7, 2026Created autopreso — Realtime speech to presentation. Let the whiteboard whiteboard itself.
- May 10, 2026Created simplewords
- May 11, 2026Most recent push to simplewords
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.