01 · Roasts
The Commit Machine Has No Off Switch
58,249 commits in a single year. That's 159 commits per day on average. The heatmap has more level-4 cells than a nuclear reactor's control panel. Peter doesn't have work-life balance — he has work-work balance.
57 Providers and Counting
CodexBar tracks 57+ AI providers. At this rate, you'll need a menu bar app to track how many menu bar apps you've shipped. The recursion is approaching critical mass.
Came Back from Retirement to Never Sleep Again
nightOwlPct: 100. That's not a night owl, that's a creature that has evolved beyond the need for daylight. Your bio says you 'came back from retirement' — retirement apparently meant switching from 23-hour workdays to 24-hour ones.
3,309 PRs in a Year
You opened 3,309 pull requests in 12 months. That's 9 PRs per day. Your git log reads less like a commit history and more like a Terms of Service agreement that GitHub never agreed to.
openclaw-pr-assets: The Skeleton at the Feast
Among 51,852 total stars and production-grade SaaS products, you have one repo that is literally just screenshots uploaded to prove PRs happened. No license. No CI. Language: Unknown. Even your asset dumps are committed with more frequency than most engineers ship features.
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% weight96S
- Consistency20% weight95S
- Quality20% weight82A
- Depth15% weight80A
- Breadth10% weight80A
- Community10% weight95S
03 · Stats
365-day commit heatmap
353 active days
Language distribution
- Swift52%
- TypeScript38%
- Objective-C4%
- JavaScript2%
- Shell2%
- Go1%
- Other1%
04 · Numbers
Owned repos
non-fork
71
Commits
last 12 months
58,249
Followers
52,304
Joined GitHub
Feb 2009
05 · Top repos
steipete /
oracle
Production-grade TypeScript CLI for AI code analysis, shipping with comprehensive test coverage, CI/CD, typed architecture, and multi-provider model support (OpenAI, Claude, Gemini, Grok). 3.2k stars, active development.
steipete /
CodexBar
Mature macOS menu bar app tracking AI coding provider usage/credits across 57+ services. Strong typed Swift codebase with tests, CI/CD, comprehensive docs, and active maintenance. Ecosystem-connecting but niche-focused tool.
steipete /
agent-scripts
Well-structured shared agent tooling with 6k stars, comprehensive portable scripts, skills framework, and extensive CI/docs for internal use across multiple projects.
steipete /
ReleaseBar
Typed full-stack open-source dashboard for GitHub release tracking. Cloudflare Worker backend with Svelte frontend, comprehensive testing (HAS_TESTS=yes, HAS_CI=yes), and well-documented architecture across design.md, ARCHITECTURE.md, STATUS.md.
steipete /
homebrew-tap
Active Homebrew tap packaging 16+ formulae and 4 casks. Well-structured Python automation (update_formula.py with 500+ LOC), comprehensive test suite, CI/CD workflows, and clear documentation. Named product with documented use patterns.
steipete /
stats-store
Privacy-first analytics service for Sparkle-enabled Mac apps. TypeScript Next.js app with Supabase backend, comprehensive docs, test coverage, and CI/CD pipeline. Shipped SaaS product with real user adoption.
steipete /
sweetlink
TypeScript agent automation tool offering Playwright-like browser control in the current tab. 155 stars, MIT licensed, with complete test+CI coverage, rich documentation (docs/, design.md, ARCHITECTURE.md), and a focused product narrative around closing agent loops.
steipete /
vox
Phone bridge connecting Twilio PSTN calls to OpenAI Realtime with agent tool support. Typed, tested, CI-enabled CLI tool with clean modular architecture serving a specific production use case.
steipete /
mcp-agentify
MCP orchestrator that routes natural-language tasks to backend MCP servers via LLM selection. Well-typed TypeScript with comprehensive tests, CI/CD, documentation (README, design docs), and a web dashboard. 678 KB codebase, ~14 months old, actively developed with MCP SDK integration.
steipete /
Markdansi
TypeScript Markdown→ANSI renderer with clean API, full test coverage, streaming support, configurable themes, and a CLI. Focused tool for terminal rendering with thoughtful wrapping/truncation logic.
steipete /
speaking
Public speaking portfolio repository curating conference appearances and past talks for established tech engineer. Well-documented with README and images, minimal code complexity—serves as professional engagement tracker rather than software artifact.
steipete /
openclaw-pr-assets
Lightweight asset repository for OpenClaw PR proof materials (screenshots/videos). Minimal scope with 30 commits in ~6 days, short-term storage utility with no tests, CI, or license.
06 · Timeline
- Feb 27, 2009Joined GitHub
- Mar 5, 2017Created speaking — Upcoming and past speaking engagements for Peter Steinberger @steipete
- May 14, 2025Created mcp-agentify — MCP orchestrator that converts MPC servers to agents.
- Jun 17, 2025Created stats-store — stats.store - Fast, open, privacy-first analytics for Sparkle.
- Jul 3, 2025Created homebrew-tap — Homebrew tap for steipete's tools
- Nov 5, 2025Created sweetlink — Connect your agent to your web app. Like playwright, but works in your current tab. Close the loop.
- 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.
- Nov 16, 2025Created Markdansi — Markdown to ANSII in TypeScript based on Micro-Mark, with support for URLs, tables, lists and more.
- Dec 13, 2025Created vox — Let's your agent run phone calls
- May 14, 2026Created ReleaseBar — Release freshness dashboard for open source maintainers
- Jul 4, 2026Created openclaw-pr-assets — Proof assets (screenshots) for OpenClaw PRs; one release tag per PR
- Jul 10, 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.