01 · Roasts
Zero-star constellation
Twelve named projects, 0 total stars: the shipping engine is running, but discovery has not boarded yet.
Systems buffet
MiniSniffer, sorlun, await-game, and terminal-rain make C/C++ look less like a language choice and more like a lifestyle.
CI selective service
careerforge and MiniSniffer brought CI; steel-helloworld and github-crawler appear to have filed for an exemption.
Horizontal builder
194 recent sampled commits across repos says this is not one heroic monolith—it's a small product factory.
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% weight68C
- Consistency20% weight65C
- Quality20% weight79B
- Depth15% weight58D
- Breadth10% weight95S
- Community10% weight25F
03 · Stats
365-day commit heatmap
92 active days
Language distribution
- C38%
- C++22%
- TypeScript19%
- Python8%
- Java5%
- JavaScript4%
- Other4%
04 · Numbers
Owned repos
non-fork
25
Commits
last 12 months
695
Followers
5
Joined GitHub
Jul 2025
05 · Top repos
finn-abel /
sorlun
A substantial, well-documented C++20 build-trace CLI with SQLite persistence, graph analysis, six ecosystem adapters, TUI, tests, and CI artifacts, but currently has no stars, forks, or demonstrated external adoption.
finn-abel /
MiniSniffer
A substantial, well-documented C11 packet analyzer with libpcap capture, IPv4/IPv6 parsing, bounded TCP/fragment reassembly, HTTP/DNS/TLS/DHCP/QUIC decoding, extensive tests, CI, sanitizers, coverage, and fuzzing, but no demonstrated adoption.
finn-abel /
await-game
A highly structured C++26 async terminal-game experiment with deterministic headless simulation, cancellation-focused architecture, broad tests, and multi-platform CI, but currently showing no adoption beyond the repository itself.
finn-abel /
careerforge
A documented, typed TypeScript monorepo implementing a substantial CareerForge product with Next.js, NestJS API/worker, PostgreSQL/pgvector, Redis queues, AI/RAG workflows, security hardening, tests, and CI, but currently has no visible adoption.
finn-abel /
shouldbe-ignitionhacks
ShouldBe is a substantial deployed fintech hackathon product with a FastAPI/React architecture, meeting-cost ledger, budget guardrails, inbound calendar email workflow, and unusually detailed security hardening.
finn-abel /
civic-bike-docks
A polished, typed civic mapping app for 1,063 Bike Share Toronto stations, with offline snapshots, live GBFS updates, validated GeoJSON, clustering, accessibility-minded UI, and computed 400 m coverage analysis; adoption is currently unproven.
finn-abel /
larp-terminal
A polished, typed Electron financial-terminal parody with deterministic market simulation, nine dockable panels, persisted workspaces, strong unit coverage, and clear documentation, but no visible adoption or CI.
finn-abel /
terminal-rain
A polished, documented C++20 terminal visualization tool with image/GIF handling, ANSI rendering, deterministic rain simulation, and substantial unit coverage, but currently showing no measurable external adoption.
finn-abel /
image81
A focused, documented Python CLI with Pillow conversion, thresholding, transparency handling, dithering, and a substantial pytest suite, but no demonstrated adoption, CI, or repository license file.
finn-abel /
hackathon-finder
A focused Python hackathon-discovery pipeline with a local dashboard, deterministic location/date screening, browser collection, LLM extraction/judging, and a substantial pytest contract suite; adoption is not yet evidenced.
finn-abel /
github-crawler
A documented, usable standard-library GitHub crawler with thoughtful retry, SSL, and GraphQL handling, but it is a same-day, zero-star two-file utility without tests, CI, typing, or licensing.
finn-abel /
steel-helloworld
A clearly documented, runnable Steel/Browser-Use Hacker News demo with careful session cleanup and setup scripts, but no tests, CI, typed-language classification, or evidence of adoption beyond a zero-star one-day project.
06 · Timeline
- Jul 2, 2025Joined GitHub
- Jun 8, 2026Created MiniSniffer — MiniSniffer: A lightweight packet sniffer and network analyzer for live traffic capture, pcap analysis, and protocol decoding (HTTP, DNS, TLS).
- Jul 2, 2026Created sorlun — Sorlun: Trace builds for most natively compiled languages and explain what rebuilt, why it rebuilt, and what a change would affect without replacing your build system.
- Jul 9, 2026Created github-crawler
- Jul 10, 2026Created image81 — image81: A CLI that converts images into ASCII art using only the characters 1 and 8.
- Jul 10, 2026Created terminal-rain — terminal-rain: Matrix-style terminal rain for generating command-line visuals, for image generation.
- Jul 13, 2026Created await-game — await-game: A tiny terminal arena game for exploring C++26 async programming, using NVIDIA stdexec.
- Aug 10, 2026Created careerforge — CareerForge: Job application management and preparation tool for students and professionals.
- Aug 22, 2026Created shouldbe-ignitionhacks — ShouldBe: a fintech dashboard that turns calendar meetings into measurable spend, flags avoidable costs, and drafts email replacements.
- Aug 28, 2026Created larp-terminal — larp-terminal: a fake financial terminal for LARPing as someone who moves markets.
- Aug 31, 2026Created civic-bike-docks — bike-docks: An interactive map of every Bike Share Toronto dock, showing live availability, and station density.
- Sep 7, 2026Created steel-helloworld — My first Steel AI browser agent
- Sep 8, 2026Created hackathon-finder — Steel AI browser agent for finding hackathons
- Sep 8, 2026Most recent push to hackathon-finder
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.