Skip to content

DavidFliesen/AWAIR

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 

Repository files navigation

AWAIR — Agentic Weekly Artificial Intelligence Report

AWAIR — Agentic Weekly Artificial Intelligence Report

Launch AWAIR   Version 1.0.0   MIT License   Local Only


AWAIR is the Agentic Weekly Artificial Intelligence Report — a privacy-first, browser-based intelligence dashboard that helps AI practitioners, researchers, and enthusiasts track the people, companies, and trends shaping the future of artificial intelligence.

Configure a personal watchlist of key AI leaders, monitor curated sources ranging from company research blogs to academic preprint servers, and generate structured weekly briefings delivered on-screen or via email. All configuration and report history is stored exclusively in your browser’s local storage — no accounts, no servers, no tracking.


✨ Features

  • 📋 Configurable Watchlist — Track key AI leaders by name, X (Twitter) handle, website, and YouTube channel. Assign priority tiers: Core, Watch, or Track.
  • 🌐 Source Library — Monitor company blogs, research feeds, newsletters, and news outlets. Toggle sources on/off at any time.
  • 🏷 Keyword Signals — Define custom topic keywords (e.g. “GPT-5”, “AI agents”, “multimodal”) that are scanned and flagged in every report.
  • 📡 On-Demand Reports — Run a full intelligence scan at any time with a single click. Reports render directly in the browser.
  • 📅 Weekly Scheduling — Configure which days of the week to auto-generate reports and at what time of day.
  • ✉ Email Delivery — Send your report to any email address via your local mail client, no SMTP configuration required.
  • 🖨 Print-Ready — Reports are formatted for clean printing or PDF export directly from the browser.
  • ⬇ Export / Import — Back up or transfer your full configuration as a portable JSON file.
  • 🔒 100% Local Storage — Zero data leaves your device. No accounts. No tracking. No third-party calls.

🚀 Getting Started

AWAIR is a single-file HTML application. No build tools, no dependencies, no server required.

▶ Run in Browser (Live Demo)

→ Launch AWAIR

Open the link above to run AWAIR instantly — no installation needed.


Run Locally

git clone https://github.com/DavidFliesen/awair.git
cd awair
open awair-v001.html   # macOS
# or double-click awair-v001.html in Windows Explorer / Linux file manager

That’s it. The app runs entirely in your browser.

Deploy to GitHub Pages

  1. Fork this repository
  2. Go to Settings → Pages
  3. Set source to main branch, root directory
  4. Your AWAIR instance will be live at https://yourusername.github.io/AWAIR/

👁 Default Watchlist

AWAIR ships with a curated starting watchlist of the most influential voices in AI:

🔴 Core Watch — The AI Big 3

Person Role X Handle
Dario Amodei CEO — Anthropic @darioadmodei
Sam Altman CEO — OpenAI @sama
Kevin Weil Chief Product Officer — OpenAI @kevinweil
Demis Hassabis CEO — Google DeepMind @demishassabis

🟡 Extended Watch

Person Role X Handle
Mark Zuckerberg CEO — Meta @zuck
Jensen Huang CEO — NVIDIA @jensenhuang
Satya Nadella CEO — Microsoft @satyanadella

🟢 Tracking

Person Role X Handle
Yann LeCun Chief AI Scientist — Meta @ylecun
Andrej Karpathy Independent AI Researcher @karpathy
Fei-Fei Li Co-Founder — World Labs / Stanford @drfeifei
Mustafa Suleyman CEO — Microsoft AI @mustafasuleyman
Elon Musk / xAI CEO — xAI / Grok @elonmusk

All people can be edited, removed, or supplemented from the Watchlist tab.


🌐 Default Sources

AWAIR pre-loads 12 curated AI intelligence sources across five categories:

Source Category
The Rundown AI Newsletter
MIT Technology Review News
Anthropic Blog Company
OpenAI Blog Company
Google DeepMind Blog Company
Hugging Face Blog Company
ArXiv AI Papers Research
AI Snake Oil (Substack) Newsletter
VentureBeat AI News
The Decoder News
Latent Space Podcast Newsletter
a16z AI News

Sources are fully configurable — toggle, remove, or add your own from the Sources tab.


🏷 Default Keywords

GPT-5 · Claude · Gemini · AI agents · multimodal · reasoning models
open source AI · AI safety · AI regulation · robotics · diffusion models · LLM

Add any keyword from the Watchlist tab and it will be scanned in every report.


🗂 App Structure

awair-v001.html        ← Complete single-file application (open to run)
README.md              ← This file
LICENSE                ← MIT License
assets/
  awair-logo.svg       ← Standalone logo asset (also used in this README)

🔒 Privacy Architecture

AWAIR was designed from the ground up with user privacy as a core principle:

  • All data lives in localStorage — watchlist, sources, keywords, settings, and report history are stored only in your browser on your device.
  • No external API calls — the app makes zero network requests to any third-party service.
  • No analytics, no telemetry — there is no usage tracking of any kind.
  • No account required — there is no login, no registration, no email verification.
  • Portable config — export your full configuration as a JSON file at any time from the Settings tab. Import it on any other device to restore your setup instantly.
  • Full reset — the Settings tab includes a one-click option to wipe all local data and restore factory defaults.

Note on scheduled email delivery: AWAIR generates a mailto: link that opens your local mail client. It does not connect to any SMTP server directly. For fully automated email delivery, a lightweight local server or browser extension relay would be needed — see the Roadmap section below.


🗺 Roadmap

Planned features for future versions:

  • v002 — RSS feed parsing for real-time source content (via CORS proxy or local server)
  • v003 — Live X/Twitter feed integration via API key (user-supplied, stored locally)
  • v004 — AI-powered report summarization using a user-configured LLM API key
  • v005 — SMTP email delivery with local credentials stored in localStorage
  • v006 — Report history archive with diff highlighting (“new since last week”)
  • v007 — Electron or PWA wrapper for true background scheduling
  • v008 — Dark/light theme toggle and custom accent colors
  • v009 — Multi-profile support (e.g. separate profiles for different focus areas)
  • v010 — Mobile-optimized layout

🛠 Built With

  • Vanilla HTML / CSS / JavaScript — zero framework dependencies
  • Google Fonts — Rajdhani, Exo 2, JetBrains Mono
  • Browser localStorage API — all persistence
  • CSS Custom Properties — full theming system
  • No npm. No build step. No bundler. Just open the file.

📄 License

MIT License — free to use, modify, and distribute. See LICENSE for details.


👤 About the Developer

David Fliesen is a Hybrid Generative AI Multimedia Developer based in Summerville, SC, with a background spanning 20+ years across multimedia production, U.S. Navy Combat Camera photojournalism, and pioneering DoD virtual agent work dating back to 2008. He recently completed the Purdue Online / Simplilearn Applied Generative AI Specialization and actively builds at the intersection of AI, creative media, and practical tooling.

David is the creator of the daily AI-generated comic strip Sisters of Summerville and develops a range of AI-powered prototypes including ComicsGen, InsightForge (an AI-powered talking BI dashboard), and AHA (an Agentic Healthcare Assistant).

🌐 Portfolio davidfliesen.github.io
💼 LinkedIn linkedin.com/in/fliesen
🐙 GitHub github.com/DavidFliesen
🎨 Sisters of Summerville sisters-of-summerville.github.io

AWAIR · Agentic Weekly AI Report · All data stored locally · No tracking · MIT License

About

AWAIR is the Agentic Weekly Artificial Intelligence Report. Privacy-first browser dashboard to track AI leaders, company blogs, research feeds, and trending topics. Generate weekly intelligence briefings on-screen or by email. All data stored locally — no accounts, no tracking.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages