01 · Roasts
The 10-Minute Repo Industrialist
kicad-footprint-generator: created AND last pushed within 10 minutes, kicadmodtree: created AND pushed within 1 second. You're out here generating repos faster than commits.
Tests? Never Heard of Her
9 out of 12 evaluated repos have HAS_TESTS=no. You've written multi-provider async LLM adapters, proto marshaling, and a full agent runtime — but apparently testing is a 2027 problem.
The 15-Week Nap
Your heatmap shows a crisp 15-week dead zone in weeks 16–31. The commit graph looks less like a developer and more like a bear going into hibernation between PCB seasons.
Speed-run Architect
Ki-Stack: 16 minutes. kiforge-kicad-bom: 81 seconds of commits. agent-sdk: entire framework in one day. You have ARCHITECTURE.md files longer than some repos have been alive.
KiCad Maximalist
You have kicad-ipc-rs, kiutils-rs, kiforge-kicad-bom, Ki-Stack, AND kicad-footprint-generator. At this point just rename your GitHub to kicad-rs and commit to the bit.
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% weight62C
- Consistency20% weight65C
- Quality20% weight72B
- Depth15% weight58D
- Breadth10% weight65C
- Community10% weight55D
03 · Stats
365-day commit heatmap
165 active days
Language distribution
- Python60%
- Jupyter Notebook22%
- Rust11%
- TypeScript5%
- Dart1%
- C++0%
- Other1%
04 · Numbers
Owned repos
non-fork
37
Commits
last 12 months
500
Followers
48
Joined GitHub
Jul 2020
05 · Top repos
Milind220 /
kicad-ipc-rs
Production-ready async-first Rust bindings for KiCad IPC API (57/57 commands implemented). Well-documented, typed, CI/CD-enabled indie library with structured multi-file architecture (1030 KB, ~20k LOC) and both async + blocking APIs.
Milind220 /
kiutils-rs
Rust-native KiCAD parser with lossless round-trip, typed APIs across 3-crate workspace. Comprehensive docs (docs/, design.md, ARCHITECTURE.md), CI, tests, and 826KB codebase with structured architecture.
Milind220 /
OpenSlate
TypeScript monorepo for multi-agent orchestration runtime with structured sessions, episodes, and tool execution. Well-documented via ARCHITECTURE.md, design.md, STATUS.md; ships tests and typed codebase; 30 commits in 6 days shows coherent sprint.
Milind220 /
agent-sdk
Early-stage Python agent framework emphasizing simplicity over abstraction. Ships with typed async LLM adapters, dependency injection, and context compaction—but 1 star, no tests, and created/pushed on same day (2026-03-01) suggest very recent release. Structured codebase with docs but lacks community validation.
Milind220 /
agent-sdk-rs
Lightweight Rust agent SDK for tool-calling with multi-provider support (Anthropic, Google, xAI). Early-stage alpha with structured src/, comprehensive docs (design.md, ARCHITECTURE.md, mdBook guide), typed code, and integration tests. 22/30 commits in ~7 days.
Milind220 /
nvim-config
Personal Neovim config with typed Lua, LSP setup, and custom theme system. Minimal documentation; no tests/CI. Active but narrow-scope portfolio piece.
Milind220 /
Milind220
Personal profile config repo with minimal content (44 KB, 2 stars). README showcases owner's other projects (KiCAD-IPC-RS, kiutils-rs, Agent SDK) but this repo itself is just config—no code, tests, or CI.
Milind220 /
sate
Personal agent skill for restaurant recommendations via thin Python API wrappers (Google Places, Open-Meteo). Well-documented SKILL.md & README for integration into Claude/Cursor, but 1 star, 4 commits in 1 day, no tests/CI, untyped Python, no license.
Milind220 /
Ki-Stack
Early-stage skill pack for KiCad automation agents with Shell scripts, structured documentation (design.md, ARCHITECTURE.md), and documented substrate/domain skills, but created/pushed same day with minimal commit history.
Milind220 /
kiforge-kicad-bom
Minimal Rust/Tauri KiCad BOM helper app with SSE-based MCP server; fresh repo (created 2026-03-04, 2 commits in 81 seconds), no tests/CI/license, typed lang + README covering tools and quick-start.
Milind220 /
kicad-footprint-generator
Empty scaffold dumped in 10 minutes with zero commits of substantive work. While it has documentation flags set, zero stars, no actual source content sampled, and created/last push within same 10-minute window mark this as a placeholder initialization.
Milind220 /
kicadmodtree
Empty scaffold repository with zero commits, no files, no documentation, and no meaningful project content. Created 2026-02-26 with immediate push, suggesting auto-initialization or bot activity.
06 · Timeline
- Jul 27, 2020Joined GitHub
- Jul 10, 2021Created Milind220 — Config files for my GitHub profile.
- Jan 14, 2024Created nvim-config — My personal Neovim config!
- Feb 18, 2026Created kicad-ipc-rs — MIT licensed, fully updated, actively maintained Rust bindings for the KiCAD IPC API
- Feb 21, 2026Created agent-sdk-rs — Lightweight (and therefore powerful) Rust agent framework inspired by Browser Use's Agent SDK
- Feb 23, 2026Created kiutils-rs — Rust-native KiCAD parser/writer - Write in Rust, convert seamlessly to KiCAD design files
- Feb 26, 2026Created kicadmodtree
- Feb 26, 2026Created kicad-footprint-generator
- Mar 1, 2026Created agent-sdk
- Mar 4, 2026Created kiforge-kicad-bom
- Mar 30, 2026Created Ki-Stack — Deep KiCad skills for coding agents
- Apr 3, 2026Created OpenSlate
- Apr 26, 2026Created sate — Your personal AI food concierge. Agent-powered restaurant recommendations that actually remember what you like.
- Apr 27, 2026Most recent push to sate
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.