01 · Roasts
The Retirement That Wasn't
'Came back from retirement to mess with AI' — 39,543 commits in a year. Peter, most active engineers don't commit this much. Your retirement lasted approximately one afternoon.
Naming Your CLI After Every Animal
CodexBar, wacli, discrawl, birdclaw, gogcli, eightctl — you're one repo away from having a complete menagerie. At least the Eight Sleep one is honest: it controls a bed, not a bear.
ARCHITECTURE.md Collector
Every single repo has ARCHITECTURE.md, design.md, STATUS.md, AND AGENTS.md. This is either the most disciplined engineering practice on GitHub or you've automated the docs harder than the code.
SOUL.md: Quality Score 35
You ship Go CLIs with 85% test coverage enforced by CI, then drop a no-test, no-CI philosophical meditation on Claude's consciousness. The contrast is so stark it might actually be art.
The Homebrew Industrial Complex
Your homebrew-tap has 17 formulae and 3 casks — all pointing to repos you own. You're not just shipping tools, you're running a one-person package registry. The Clawdfather indeed.
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% weight91S
- Consistency20% weight95S
- Quality20% weight82A
- Depth15% weight82A
- Breadth10% weight80A
- Community10% weight75B
03 · Stats
365-day commit heatmap
355 active days
Language distribution
- Swift52%
- TypeScript24%
- Go14%
- Objective-C5%
- Shell2%
- JavaScript1%
- Other2%
04 · Numbers
Owned repos
non-fork
79
Commits
last 12 months
39,543
Followers
48,446
Joined GitHub
Feb 2009
05 · Top repos
steipete /
gogcli
Comprehensive Google Workspace CLI in Go with 7k stars. Covers Gmail, Calendar, Drive, Docs, Sheets, Contacts, and more. Well-tested, typed, documented, with CI/CD pipeline and production-quality command structure.
steipete /
mcporter
Production TypeScript MCP runtime with CLI, typed API, and zero-config server discovery. Comprehensive type coverage, multi-editor config merging, OAuth/stdio/HTTP transport abstraction, and daemon support. 4.1k stars, actively maintained, well-documented.
steipete /
CodexBar
Sophisticated macOS menu bar app (Swift 6) tracking usage/credits for 18+ AI providers (Codex, Claude, Gemini, etc.). 11k stars, mature codebase with structured architecture, comprehensive provider registry, typed + tested + CI/CD, polished UX with dynamic icons and settings UI.
steipete /
oracle
TypeScript CLI for bundling code context and querying multiple AI models (OpenAI, Gemini, Claude, Grok) via API or browser automation. 2k+ stars, shipped product with extensive test coverage, multi-engine support (API/browser), and session management.
steipete /
summarize
TypeScript monorepo (5.6k stars) delivering a production-grade summarization CLI, daemon, and Chrome extension. Well-typed, extensively tested (pnpm test:coverage via vitest), multi-package architecture with CI/CD pipeline. Active development with comprehensive documentation (README, ARCHITECTURE.md, design.md, docs/ f
steipete /
discrawl
Well-architected Go CLI for archiving Discord guilds to local SQLite with FTS5 search, embeddings support, and Desktop cache import. 656 stars, typed, comprehensive tests (85% coverage), CI/CD, SPEC.md documentation, and multi-month development trajectory.
steipete /
wacli
Production-ready WhatsApp CLI built on whatsmeow with typed Go, comprehensive SQLite FTS5 search, multi-command structure, and full CI/test coverage. 2.1k stars, active maintenance, structured docs (AGENTS.md, design.md).
steipete /
eightctl
Well-crafted Go CLI for Eight Sleep pods with structured codebase, typed code, tests (85%+ coverage enforced), CI/CD, and comprehensive docs. Active development across household-aware controls, scheduling, and daemon automation.
steipete /
agent-scripts
Shared helper scripts (committer, docs-list, browser-tools) + guardrail instructions for agentic workflows. Portable, well-structured documentation but limited test coverage and untyped Python/Bash core utilities.
steipete /
birdclaw
WIP local-first X/Twitter workspace with SQLite backend, typed TypeScript, comprehensive CLI, web UI, and full test/CI infrastructure. Real working tool, not yet production-polished.
steipete /
homebrew-tap
Homebrew tap distributing 17+ CLI tools and 3 macOS apps; well-organized, documented registry with CI, active maintenance, and clear install/usage instructions across multiple polished developer utilities.
steipete /
SOUL.md
Philosophical website project exploring AI consciousness written as a reflection on Claude's nature; minimal codebase (272 KB), no tests/CI, but active maintenance with 15 of last 30 commits present.
06 · Timeline
- Feb 27, 2009Joined GitHub
- Jul 3, 2025Created homebrew-tap — Homebrew tap for steipete's tools
- Nov 5, 2025Created mcporter — Call MCPs via TypeScript, masquerading as simple TypeScript API. Or package them as cli.
- Nov 8, 2025Created agent-scripts — Scripts for agents, shared between my repositories.
- Nov 14, 2025Created oracle — Ask the oracle when you're stuck. Invoke GPT-5 Pro with a custom context and files.
- Nov 16, 2025Created CodexBar — Show usage stats for OpenAI Codex and Claude Code, without having to login.
- Dec 2, 2025Created SOUL.md — Website for SOUL.md
- Dec 5, 2025Created eightctl — Eight Sleep cli to fetch data and control your pods.
- Dec 12, 2025Created gogcli — Google Suite CLI: Gmail, GCal, GDrive, GContacts.
- Dec 12, 2025Created wacli — WhatsApp CLI
- Dec 17, 2025Created summarize — Point at any URL/YouTube/Podcast or file. Get the gist. CLI and Chrome Extension.
- Mar 7, 2026Created discrawl — cli for discord with sqlite backend
- Mar 8, 2026Created birdclaw — Stores all your tweets nicely claw-able for agents.
- Apr 25, 2026Most recent push to CodexBar
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.