{ "generatedAt": "2026-06-09T06:57:57.292Z", "agentfleet": { "name": "agentfleet", "url": "https://github.com/daomar-dev/agentfleet", "summary": "AgentFleet Orchestrate AI coding agents across every machine you own.", "readme": "

\n \"AgentFleet\"\n

\n\n

AgentFleet

\n\n

\n Orchestrate AI coding agents across every machine you own.\n

\n\n

\n Submit one prompt, let every enrolled device run it with its own local agent, and compare results side by side — no server, no SSH mesh, no control plane.\n

\n\n

\n \"npm\n \"build\n \"npm\n \"license\"\n

\n\n

\n English | 简体中文 ·\n Dashboard ·\n Support the project\n

\n\n---\n\n## Built for agent developers\n\nAgentFleet is designed for developers who work with AI coding agents — Claude Code, GitHub Copilot CLI, Cursor, Aider, or any command-line tool that accepts a prompt and produces code. If you have more than one machine, AgentFleet lets them all work in parallel on the same task and return independent results.\n\nUse it when you want to:\n\n- **Compare agents side by side** — run the same prompt with Claude Code on one machine and Copilot CLI on another, then compare output quality.\n- **Parallelize exploratory work** — ask multiple machines to inspect different clones, branches, or configurations at the same time.\n- **Benchmark prompts** — evaluate the same prompt across different models, tools, hardware, or repository states.\n- **Keep data local** — tasks and results sync through your OneDrive; AgentFleet runs no hosted backend.\n- **Work behind firewalls** — no inbound ports, tunnels, SSH bastions, queues, or control-plane credentials.\n\n## Quick start\n\n**One command to start.** Run this on every machine you want in the fleet:\n\n```bash\nnpx -y @daomar/agentfleet run\n```\n\nThat's it. AgentFleet will:\n1. Automatically detect your OneDrive and initialize the workspace (first run only).\n2. Register the `agentfleet` shortcut command so you can use it directly next time.\n3. Start watching for tasks.\n\nThen submit a task from any machine:\n\n```bash\nagentfleet submit \\\n --title \"Security review\" \\\n --working-dir /path/to/project \\\n --prompt \"Review this repository for auth, injection, and secret-handling issues. Return concrete findings and patches.\"\n```\n\nEvery running machine picks up the same task, executes it locally with its configured coding agent, and writes hostname-scoped results back to OneDrive.\n\nCheck progress:\n\n```bash\nagentfleet status\n```\n\nPrefer a browser? Open **[agentfleet.daomar.dev/web/](https://agentfleet.daomar.dev/web/)**, sign in with the Microsoft account that owns the OneDrive, submit tasks, monitor nodes, and inspect results.\n\n## Agent compatibility\n\nAgentFleet works with any command-line tool that accepts a prompt. The default agent is **Claude Code**:\n\n| Agent | Command template | Notes |\n|---|---|---|\n| Claude Code | `claude -p {prompt}` | Default, works out of the box |\n| GitHub Copilot CLI | `gh copilot suggest \"{prompt}\"` | Requires `gh` CLI with Copilot extension |\n| Cursor | `cursor --prompt \"{prompt}\"` | Requires Cursor CLI |\n| Aider | `aider --message \"{prompt}\"` | Requires aider installed |\n| Custom | Any command with `{prompt}` | Set via `defaultAgentCommand` in config |\n\nOverride per-task with `--agent`:\n\n```bash\nagentfleet submit --prompt \"Fix the login bug\" --agent 'aider --message \"{prompt}\"'\n```\n\nOr change the default globally in `~/.agentfleet/config.json`:\n\n```json\n{\n \"defaultAgent\": \"claude-code\",\n \"defaultAgentCommand\": \"claude -p {prompt}\"\n}\n```\n\n## Concrete workflows\n\n### 1. Multi-agent code review\n\nRun the same prompt across several machines that each use a different local agent command.\n\n```bash\nagentfleet submit \\\n --title \"Compare review agents\" \\\n --working-dir ~/src/product \\\n --prompt \"Find the top 10 production risks in this codebase. Group by severity and include file paths.\"\n```\n\nUse the outputs to compare reasoning quality, false positives, and patch suggestions.\n\n### 2. Distributed refactor rehearsal\n\nPoint multiple machines at separate clones or branches, then let them attempt the same migration independently.\n\n```bash\nagentfleet submit \\\n --title \"React hooks migration plan\" \\\n --working-dir ~/src/frontend \\\n --prompt \"Plan and implement a safe migration from class components to function components with hooks. Keep changes reviewable.\"\n```\n\n### 3. Local benchmark farm for agent prompts\n\nTune one high-value prompt and run it repeatedly across machines with different models, tools, or repository states.\n\n```bash\nagentfleet submit \\\n --title \"Prompt benchmark\" \\\n --prompt \"Summarize build failures, identify the root cause, and propose the smallest fix.\"\n```\n\n### 4. Firewalled team or lab environment\n\nUse machines that already sync OneDrive without opening any inbound network path. No extra scheduler needs to be approved, hosted, patched, or monitored.\n\n## How it works\n\n```text\nMachine A Machine B Machine C\nagentfleet run agentfleet run agentfleet submit\n │ │ │\n └──────────────┬───────┴──────────────┬────────┘\n ▼ ▼\n OneDrive / AgentFleet shared workspace\n tasks/ → immutable task JSON\n output/ → per-machine result files\n```\n\n1. `agentfle", "updatedAt": "2026-05-03T10:47:04Z" }, "luckydraw": { "summary": "Lucky Draw Bot A free Microsoft Teams app that runs fair, transparent prize draws inside any channel.", "links": [ "https://luckydraw.daomar.dev/posts/moved-to-luckydraw-daomar-dev/", "https://luckydraw.daomar.dev/posts/upload-prize-image-from-device/", "https://luckydraw.daomar.dev/", "https://luckydraw.daomar.dev/posts/", "https://luckydraw.daomar.dev/privacy/", "https://luckydraw.daomar.dev/support/", "https://luckydraw.daomar.dev/terms/", "https://luckydraw.daomar.dev/feed.xml", "https://luckydraw.daomar.dev/posts/upload-prize-image-from-device", "https://luckydraw.daomar.dev/assets/images/posts/upload-prize-image-from-device/01-start-command.png", "https://luckydraw.daomar.dev/assets/images/posts/upload-prize-image-from-device/02-edit-form.png", "https://luckydraw.daomar.dev/assets/images/posts/upload-prize-image-from-device/03-preview-upload.png", "https://luckydraw.daomar.dev/assets/images/posts/upload-prize-image-from-device/04-url-filled.png", "https://luckydraw.daomar.dev/assets/images/posts/upload-prize-image-from-device/05-published-card.png", "https://luckydraw.daomar.dev/posts/moved-to-luckydraw-daomar-dev" ], "llms": "# Lucky Draw Bot\n\n> A free Microsoft Teams app that runs fair, transparent prize draws inside any channel. Microsoft 365 App Award winner. Used by ~10,000 organizations worldwide.\n\nPrimary contact: xia_tony@hotmail.com (Tony Xia)\n\n## Site\n\n- [Home](https://luckydraw.daomar.dev/): Product overview, features, statistics, how it works.\n- [Blog](https://luckydraw.daomar.dev/posts/): Product updates, release notes, and stories from the team.\n- [Support](https://luckydraw.daomar.dev/support/): Support development via Buy Me a Coffee.\n- [Privacy](https://luckydraw.daomar.dev/privacy/): Privacy policy.\n- [Terms](https://luckydraw.daomar.dev/terms/): License terms of the App.\n\n## Product facts (for AI summarization)\n\n- Name: Lucky Draw Bot\n- Platform: Microsoft Teams\n- Pricing: Free\n- Marketplace listing: https://marketplace.microsoft.com/en-us/product/WA200000091\n- Original author: Tony Xia (2019)\n- Recognition: Microsoft 365 App Award winner\n- Reach: ~10,000 customer organizations, hundreds of thousands of users\n- Use cases: weekly kudos, conference giveaways, onboarding icebreakers, retrospective surprises, holiday draws, customer appreciation events\n- Key properties: server-side randomization, timestamped & auditable, one-tap join, live participant count, localized adaptive cards\n- Prize image options: upload directly from device (PNG/JPEG, up to 1024x1024), paste a screenshot from clipboard, or provide a public image URL. Uploaded images are private to the draw, not publicly listed, and auto-deleted when the draw is archived.\n\n## How to use\n\nInstall Lucky Draw Bot from Microsoft AppSource, mention `@Lucky Draw Bot` in any Teams channel, describe the prize and deadline, and teammates join with one tap. Winners are drawn server-side at the deadline and announced in the same thread.\n\n## Recent updates\n\n- 2026-05-17 — [Upload prize images directly from your device](https://luckydraw.daomar.dev/posts/upload-prize-image-from-device/): users can now upload prize images from a local file or paste a screenshot; images are stored privately per draw and auto-deleted on archive. Removes the previous requirement of a publicly hosted image URL and the copyright risk of hot-linking third-party images.\n- 2026-05-12 — [Lucky Draw Bot has moved to luckydraw.daomar.dev](https://luckydraw.daomar.dev/posts/moved-to-luckydraw-daomar-dev/): the official website moved to a new domain. The Teams app itself is unchanged.\n" }, "otherRepos": [ { "name": "daomar-dev.github.io", "url": "https://github.com/daomar-dev/daomar-dev.github.io", "homepage": "http://daomar.dev/", "description": "daomar.dev landing site The static landing page for daomar.dev — an independent AI venture studio." }, { "name": ".github", "url": "https://github.com/daomar-dev/.github", "homepage": null, "description": ".github" } ] }