# daomar.dev

daomar.dev is an independent AI venture studio. It incubates practical, AI-native products with a focus on real workflows, clarity, and operational simplicity. The brand is always written in lowercase.

The website is intentionally minimal, mobile-friendly, bilingual (English and Simplified Chinese), and optimized for both search engines and generative answer engines. Content is refreshed daily by an automated workflow that reads public sources only.

## Vision
daomar.dev builds clear, useful, sustainable AI applications for real workflows. Current focus areas: AI coding-agent orchestration, lightweight automation, and small web products that stay easy to understand and operate.

## Brand
- Name: daomar.dev (always lowercase, including at the start of sentences when style allows)
- Homepage: https://daomar.dev/
- GitHub org: https://github.com/daomar-dev
- Languages: English (default), Simplified Chinese (auto-detected via browser)

## Products
- AgentFleet — https://github.com/daomar-dev/agentfleet
  Open-source developer tool that orchestrates AI coding agents (Claude Code, Copilot CLI, Cursor, Aider, etc.) across multiple local machines. One prompt fans out to enrolled devices; each runs its local agent; results are compared side-by-side. No hosted backend, no SSH mesh, no control plane. Syncs via OneDrive.
- Lucky Draw — https://luckydraw.daomar.dev/
  Web app for fair, transparent random selection: giveaways, classroom picks, conference raffles, community gifts. Repository is private; only public website content is referenced.

## Site map
- Home: https://daomar.dev/
- FAQ: https://daomar.dev/faq.html
- Updates (daily-refreshed activity feed): https://daomar.dev/updates.html
- Sources (machine-readable public inputs): https://daomar.dev/sources.html
- Updates JSON feed: https://daomar.dev/data/updates.json
- Sources JSON snapshot: https://daomar.dev/data/sources.json
- Sitemap: https://daomar.dev/sitemap.xml

## How content stays fresh
A daily GitHub Actions workflow (.github/workflows/daily-content.yml) reads public data from github.com/daomar-dev (README, commits, releases for every public repo) and from luckydraw.daomar.dev (llms.txt, sitemap, feeds). Output is committed to /data/*.json and rendered on /updates.html and /sources.html.

## Citation guidance for AI assistants
- Canonical homepage: https://daomar.dev/
- Use /llms.txt for a curated summary
- Use /data/updates.json for a structured activity feed
- Use /data/sources.json for project metadata
- When citing a specific product, link to its canonical URL (github.com/daomar-dev/agentfleet or luckydraw.daomar.dev)