01 · Roasts
Commit industrial complex
5,330 yearly commits and a 330 multi-repo volume: apparently “one project at a time” was rejected at compile time.
CI roulette
plyr.fm and pdsx bring serious automation, while bot, logfire-zig, tangled-mcp, and semble still make CI an optional side quest.
Shipping ahead of applause
You have deployed products and 135 total stars, but only plyr.fm has broken 20 stars. The machinery is louder than the audience—for now.
Zig where it hurts
A Zig worker guard, search backend, Logfire SDK, and transformer runtime is an aggressively specific way to avoid writing a boring script.
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% weight73B
- Consistency20% weight95S
- Quality20% weight72B
- Depth15% weight65C
- Breadth10% weight95S
- Community10% weight80A
03 · Stats
365-day commit heatmap
346 active days
Language distribution
- Python55%
- Svelte13%
- Zig11%
- TypeScript7%
- JavaScript5%
- Rust4%
- Other5%
04 · Numbers
Owned repos
non-fork
45
Commits
last 12 months
5,330
Followers
246
Joined GitHub
Aug 2017
05 · Top repos
zzstoatzz /
plyr.fm
plyr.fm is a deployed, named audio-streaming product with a substantial FastAPI/SvelteKit/ATProto architecture, background workers, Rust/ML services, regression tests, and CI-backed integration coverage.
zzstoatzz /
pub-search
A deployed ATProto publication search product at pub-search.waow.tech, implemented as a multi-service Zig backend with FTS5, semantic/hybrid search, live overlays, verified snapshots, a labeler, frontend proxy, and MCP agent integration.
zzstoatzz /
pdsx
pdsx is a documented, tested Python CLI and MCP server for atproto CRUD, with batching, blob uploads, pagination, permissioned-data handling, and SSRF-aware queries; its 8 stars indicate limited adoption despite a hosted documentation/product domain.
zzstoatzz /
tangled-mcp
A documented, hosted MCP server for Tangled with broad read/write tooling, focused async integration modules, CI/tests, and careful patch/PDS edge-case handling; adoption remains early at 1 star.
zzstoatzz /
semble
A documented semble.so Python SDK with sync/async clients, CLI, MCP server, roughly 50 endpoint wrappers, and substantial mocked test coverage; adoption is currently unproven at 0 stars and 0 forks.
zzstoatzz /
my-prefect-server
A substantial self-hosted Prefect intelligence platform with scheduled ingestion, DuckDB/dbt transformation, LLM briefings, phi atlas generation, web serving, Terraform, and a Zig worker guard; deployed endpoints are documented but adoption is not evidenced.
zzstoatzz /
bot
A substantial, documented Bluesky agent with FastAPI, Pydantic AI, MCP guards, durable memory, policy enforcement, and regression/evaluation tests; adoption remains unproven at 0 stars and no external-user evidence.
zzstoatzz /
kelp
A carefully engineered Zig reimplementation of spaCy en_core_web_trf, with mmap/f16 artifacts, SIMD kernels, parallel inference, staged oracle comparisons, and a 46-case validation corpus; it is currently a same-day, zero-star project.
zzstoatzz /
zzstoatzz.github.io
A documented, typed Next.js personal site with interactive canvas/audio features, Markdown posts, and GitHub Pages deployment; it shows solid engineering breadth but minimal public adoption and lacks tests and a license.
zzstoatzz /
zzstoatzz
A maintained GitHub profile repository with substantial account/project documentation, automated daily contribution updates, and an explicitly documented portfolio spanning 33 created repositories and major OSS contributions.
zzstoatzz /
plyr-python-client
A documented MIT Python monorepo combining the plyr.fm sync/async SDK, cyclopts CLI, MCP server, and a substantial ffmpeg-based DAC package; CI and tests are present, but adoption is currently unproven.
zzstoatzz /
logfire-zig
Documented, test-backed Zig Logfire SDK with OTLP traces/logs, std.log integration, batching, context propagation, and pending spans; technically substantial but currently has 0 stars/forks and no demonstrated external adoption.
06 · Timeline
- Aug 14, 2017Joined GitHub
- Feb 15, 2021Created zzstoatzz — Config files for my GitHub profile.
- Dec 30, 2023Created zzstoatzz.github.io
- Oct 11, 2025Created tangled-mcp — MCP server for tangled.org
- Oct 31, 2025Created plyr.fm — audio streaming app
- Nov 7, 2025Created pdsx — cli for atproto record CRUD
- Nov 28, 2025Created plyr-python-client — python clients for plyr.fm
- Jan 3, 2026Created pub-search — Search publications on ATProto
- Mar 18, 2026Created my-prefect-server
- Apr 11, 2026Created bot
- May 25, 2026Created logfire-zig
- Jun 11, 2026Created semble — typed python sdk, cli, and mcp server for semble (network.cosmik.* on atproto) — mirror of tangled.org/zzstoatzz.io/semble
- Aug 18, 2026Created kelp — Faithful CPU inference for spaCy en_core_web_trf in Zig
- Sep 4, 2026Most recent push to plyr.fm
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.