Try a demo · OSS contributions · Qiita · Sponsor
I'm shoka, a software engineer in Shizuoka, Japan. I build tools that connect vehicle data and AI, and contribute focused documentation improvements to open source.
静岡を拠点に、自動車 × AIのツール開発と、OSSのドキュメント改善に取り組んでいます。
obd2-mcp connects MCP clients to OBD-II trouble codes, live sensors, and freeze-frame data. The included mock vehicle lets you try the tools without a car or adapter.
automotive-ai-skills provides 16 structured AI workflows for diagnostics, maintenance, used-car inspection, and more, with English and Japanese documentation.
The reproducible mock demo reads a stored P0420, its freeze frame, and current sensor values. It includes observed program output and a script you can run without an AI subscription. All data in that demo is simulated.
I work on small, verifiable improvements: understanding the issue, checking for existing work, and validating the change.
| Project | Contribution | Status checked 2026-10-29 |
|---|---|---|
| VS Code documentation | Corrected wording in the settings UX guidelines — #10552 | Merged |
| pytest | Documented platform-conditional warning filters, with a tested example — #15274 | Open PR · not yet merged |
These are individual contributions, not maintainer roles. See the linked PRs for current status, validation details, and disclosed AI assistance.
All VS Code docs PRs · All pytest PRs
Follow @shoka-jp for automotive AI tools, practical examples, and open-source improvements. Tried a project? Reproducible feedback, documentation fixes, and translations are welcome.
If the projects are useful to you, sponsor my open-source work. Support helps me work on maintenance, examples, and clearer documentation. Read about the projects you support →
- 作っているもの： obd2-mcpで車両データをAIにつなぎ、automotive-ai-skillsで診断・整備・中古車チェックなどの対話を整理しています。
- OSSへの貢献： VS Codeの設定ガイド修正はマージ済み。pytestには、条件付き警告フィルターの説明を提案しています（2026-10-29確認、未マージ）。
- まず試す： 車・アダプタ・AIの契約なしで試せるモックデモ。実車を扱う前に、各プロジェクトの安全上の注意と制約も確認してください。
- 活動を応援： フォロー、使ってみた感想、翻訳・説明の修正を歓迎します。スポンサーによるご支援も受け付けています。