01 · Roasts
Ghost Town Portfolio
9 total stars across 28 repos. Even your most polished work — sentinel-oss's 67-model Postgres schema, paper-reader's 18-paper golden corpus — has zero stars, zero forks. You're building for an audience of one.
Commit Burstor
Your heatmap looks like an EKG in cardiac arrest: 14 weeks of flatline, then a frenzy of 4s, then silence again. otherworld-archive shipped its entire ed25519 cryptography stack in 12 minutes across 3 commits. Sustained effort is a foreign concept.
Documentation Cosplayer
connected-minds has ARCHITECTURE.md, STATUS.md, VISION.md, open-questions.md, and thesis.md — and exactly zero lines of runnable code. You've written more docs about the protocol than it would take to just build the protocol.
The Quality=0 Incident
drp_02 is your deepest repo — 8 DB tables, full state machine, Expo + tRPC — and its per-repo quality score is 0. Whatever happened there, you managed to score a zero on quality in your best project.
75 PRs, 11 Followers
You've submitted 75 pull requests this year — a genuinely impressive external contribution rate — and somehow have 11 followers to show for it. You're giving more than you're getting, and nobody noticed.
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% weight65C
- Quality20% weight72B
- Depth15% weight65C
- Breadth10% weight65C
- Community10% weight50D
03 · Stats
365-day commit heatmap
115 active days
Language distribution
- TypeScript70%
- C++15%
- HTML9%
- CSS1%
- JavaScript1%
- Go1%
- Other3%
04 · Numbers
Owned repos
non-fork
27
Commits
last 12 months
1,289
Followers
11
Joined GitHub
Feb 2022
05 · Top repos
gong8 /
drp_02
Polished full-stack meetup app: Expo/React Native + Fastify/tRPC monorepo with comprehensive schema-driven API, end-to-end typing, CI/tests, typed code, docs, and production-grade architecture despite zero external visibility.
gong8 /
sentinel-oss
Self-hosted governance layer for MCP servers with policy engine, audit logging, and multi-transport proxy. Early-stage project (2 days old) with strong architectural foundation: 67-model Postgres schema, TypeScript, comprehensive CI/tests, detailed specs. Achieves active portfolio baseline due to non-trivial scope (~17
gong8 /
otherworld-archive
TypeScript protocol implementation for a multi-agent intent-negotiation system with cryptographic signing, LLM-backed minds, and a working Glass House household simulator with web UI. Well-documented with tests and structured packages, but nascent (3 commits in 12 minutes).
gong8 /
otherworld
Otherworld fabric: Go-based agent negotiation engine with LLM-powered voices, orchestrator routing, PostgreSQL transcript record. Early-stage but architecturally ambitious system for multi-party settlement negotiation.
gong8 /
paper-reader
A sophisticated academic PDF parser that extracts semantic structure and rebuilds mathematics from glyph geometry, paired with a React UI showing dual linked views. Lacks stars/adoption but represents sustained technical work on a specific domain problem.
gong8 /
3Roads
Quiz bowl question generator monorepo with API server, React frontend, and MCP integration. Typed TypeScript, structured multi-package layout, Prisma ORM, tests present but CI missing. 0 stars; experimental-stage personal project.
gong8 /
sentinel-web
Static HTML landing page for Sentinel project. One-file site with no build, testing, or CI infrastructure. Lightweight marketing asset with minimal scope.
gong8 /
connected-minds
Early-stage protocol thesis repository with comprehensive design documentation (thesis.md, open-questions.md, ARCHITECTURE.md) and clear contributing guidelines, but no implementation, tests, CI, or executable code—pure research seed.
06 · Timeline
- Feb 1, 2022Joined GitHub
- Jan 17, 2026Created sentinel-web — sentinel landing page
- Mar 20, 2026Created 3Roads — trivial quizbowl-style question generated, uses claude-cli
- May 26, 2026Created drp_02
- Jun 10, 2026Created otherworld-archive
- Jun 11, 2026Created otherworld
- Aug 18, 2026Created sentinel-oss — Self-hosted governance layer for MCP servers: policy engine, audit log, and multi-transport proxy between AI agents and the tools they call.
- Aug 25, 2026Created paper-reader — An academic PDF reader: the printed page and a reflowed reading view, side by side, linked by one model parsed in your browser.
- Aug 26, 2026Created connected-minds — An open protocol and shared infrastructure for independently reasoning agents to discover, communicate and coordinate.
- Aug 26, 2026Most recent push to connected-minds
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.