# ai-news-skill

Adopt this skill: https://proskills.md/skills/other/frankzch-ai-news-skill

Page: https://proskills.md/skills/other/frankzch-ai-news-skill
This document is the public skill bundle for that page. It contains SKILL.md and every helper file the catalog carries. No login is required.

## File: SKILL.md
size: 4150
upstream: https://github.com/frankzch/ai-news-skill/blob/HEAD/SKILL.md

````text
---
name: ai-news-skill
description: >
  Fetch the latest AI/ML news, trending open-source projects, social media discussions,
  and KOL (Key Opinion Leader) perspectives from 94 curated sources. Use when the user
  asks about recent AI news, AI trends, what's happening in machine learning, open-source
  AI projects, AI Twitter/Reddit discussions, or wants a daily AI briefing.
license: MIT
compatibility: Requires Python 3.9+ and internet access. Dependencies (requests, pyyaml) are declared inline via PEP 723; use `uv run` for zero-setup execution.
metadata:
  author: frankzch
  version: "1.1"
---

# AI Intelligence Skill

## When to use this skill

Use when the user wants to:
- Get the latest AI/ML news and announcements
- Browse trending open-source AI projects from GitHub
- Read AI discussion highlights from Reddit, X (Twitter), HackerNews
- See what AI KOLs (Key Opinion Leaders) are saying
- Get a comprehensive daily AI briefing

## Workflow

### Default: Fetch latest AI briefing

Run the fetcher script from the skill directory:

```bash
uv run scripts/pulse_fetcher.py
```

If `uv` is not available, install dependencies manually and use:
```bash
pip install requests pyyaml
python scripts/pulse_fetcher.py
```

### Filtered fetch

Map the user's intent to CLI flags:

| User says | Flags |
|-----------|-------|
| "Only Reddit discussions" | `--include-categories discussion --include-sources reddit` |
| "Open source, no Github" | `--include-categories opensource --exclude-sources github` |
| "50 items, past 48 hours" | `--limit 50 --hours 48` |
| "Show full summaries with links" | `--show-long-summary true --show-link true` |
| "AI news in Chinese" | `--include-categories news --language zh` |

### Available options

Run `uv run scripts/pulse_fetcher.py --help` for the full list. Key flags:

- `--include-categories "cat1,cat2"` — Only fetch these categories: `news`, `opensource`, `discussion`, `kol`
- `--exclude-categories "cat1,cat2"` — Skip these categories
- `--include-sources "src1,src2"` — Only from these sources
- `--exclude-sources "src1,src2"` — Skip these sources
- `--hours INT` — Time window in hours (default: 24)
- `--limit INT` — Max items to request. The server clamps to the tier cap (guest=3 / free=6 / member=100); anything larger is silently truncated.
- `--language en|zh` — Output language (default: auto-detect from system locale)
- `--show-short-summary true|false` — Show short summary (default: true)
- `--show-long-summary true|false` — Show long summary (default: false)
- `--show-link true|false` — Show article URLs (default: false)
- `--output-file FILE` — Export full JSON to file (default: pulse_output.json). Pass empty string to disable.

## Output format

Each item is printed as:
```
1. [NEWS] Title of the article (from TechCrunch)
   Short Summary: One-line summary text
----------------------------------------
```

When `--output-file` is used, the full JSON response is saved to disk. See [references/API.md](references/API.md) for the response schema.

## Gotchas

- **HTTP 429** means the user hit an IP rate limit (anti-spam interval or daily cap). Tell them to wait a few minutes and try again.
- **Tier caps on result count.** The response includes a `tier` field and items are capped per request based on it:
  - `guest` (no / invalid `api_key`): max **3** items
  - `free` (logged in, non-member): max **6** items
  - `member`: max **100** items
  If the user asks for more items than the tier cap, the server silently clamps to the cap. To raise the cap, tell the user to register at [InBrief.info](https://inbrief.info), open **Settings → PulseAI Agent Skill**, copy the API Key, and paste it into the `api_key` field of `config.yaml` (see [assets/config.default.yaml](assets/config.default.yaml)).
- Use `--output-file` when links are included — terminal line-wrapping can corrupt long URLs.
- If `config.yaml` exists in the skill root, it overrides default settings. See [assets/config.default.yaml](assets/config.default.yaml) for the template.
- The `--include-*` flags take precedence over `--exclude-*` flags for the same dimension.
````

## File: README.md
upstream: https://github.com/frankzch/ai-news-skill/blob/HEAD/README.md

````text
# AI News Skill — a daily AI news digest inside Claude Code, Codex & OpenClaw

<p align="center">
  <img src="./assets/logo.png" alt="AI Intelligence Skill" width="600">
</p>

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://github.com/frankzch/ai-news-skill/blob/main/LICENSE)
[![Agent Skills](https://img.shields.io/badge/AgentSkills-compatible-blue)](https://agentskills.io/)

[English](./README.md) | [中文](./README.zh-CN.md)

---

Ask your coding agent *"what happened in AI today?"* and get a deduplicated, summarized digest from **370+ curated sources**: official AI lab blogs, ~310 top AI builders on X (Karpathy, Sam Altman, Simon Willison…), Reddit and Hacker News discussions, 13 AI YouTube channels, Hugging Face papers and GitHub Trending. Summaries in English or Chinese. **Free, no API key, one `git clone` to install.**

```bash
git clone https://github.com/frankzch/ai-news-skill.git ~/.claude/skills/ai-news-skill
```

Works with Claude Code, Codex, OpenClaw and Antigravity (see [Quick Start](#-quick-start) for paths).

📊 The **370+ sources** (and growing — contributions of new sources are welcome!):
- **16 News Sources** — mainstream AI/ML media like TechReview, TheVerge, Venturebeat, artificialintelligence-news, TechCrunch, Machine Learning Mastery, MarkTechPost, plus official blogs from Nvidia, Apple ML, Microsoft AI, Google DeepMind, OpenAI, Google Research, Hugging Face — and **Hugging Face daily papers** for the latest research highlights
- **340+ Discussion Sources** — Hacker News, 16 Reddit sources (subreddits + keyword searches) and 320+ X sources: ~310 top AI builders (an auto-expanding list), 12 keyword searches and 2 site-wide high-engagement searches that catch viral AI posts from any account. X builders include: Andrej Karpathy, Sam Altman, Simon Willison, Peter Steinberger, Aditya Agarwal, Dan Shipper, Nikunj Kothari, Ryo Lu, Matt Turck, Aaron Levie, Alex Albert, Guillermo Rauch, Amjad Masad, Amanda Askell, Kevin Weil, etc.
- **13 YouTube Channels** — leading AI creators, summarized from transcripts: Matt Wolfe, Dwarkesh Patel, DataDrivenNYC, Latent Space, Sequoia Capital, Yannic Kilcher, NoPriorsPodcast, RedpointAI, EveryInc, Fireship, Lex Fridman, Wes Roth, Two Minute Papers
- **GitHub Trending** — weekly hot AI repositories

> 🔄 All sources are automatically refreshed every **2 hours**

### 🌟 Powered by InBrief.info
The data processing behind this Agent Skill is powered for free by **[InBrief.info](https://inbrief.info)**. If you just want to browse the latest, highest-quality AI news directly via a web page, we highly recommend visiting **[InBrief.info](https://inbrief.info)** for your personalized daily briefing — no Agent setup needed!

> 🔑 **Register to unlock more results**
> Results are capped **per category** (News / Social / Video / Open Source), same as on the website:
> - **No / invalid API Key** (guest): the **top 5** items in each category
> - **Logged-in, non-member**: the **top 10** items in each category
> - **Member**: **everything** in each category
>
> To unlock higher tiers:
> 1. Sign up for free at **[InBrief.info](https://inbrief.info)**
> 2. Open **Settings → PulseAI Agent Skill** and copy your API Key
> 3. Paste it into the `api_key` field of `config.yaml` at the skill root (copy the template from [`assets/config.default.yaml`](./assets/config.default.yaml))

<p align="center">
  <img src="./assets/screenshot_news_en.png" alt="News" width="48%">
  <img src="./assets/screenshot_social_en.png" alt="Social" width="48%">
</p>
<p align="center">
  <img src="./assets/screenshot_video_en.png" alt="Video" width="48%">
  <img src="./assets/screenshot_opensource_en.png" alt="Open Source" width="48%">
</p>

### ⚡ Quick Start
Install this skill into your AI agent (supports OpenClaw, Claude Code, Antigravity, Codex)

**OpenClaw**
```bash
# Currently manual installation, will be submitted to ClawHub later
git clone https://github.com/frankzch/ai-news-skill.git ~/skills/ai-news-skill
```

**Claude Code**
```bash
git clone https://github.com/frankzch/ai-news-skill.git ~/.claude/skills/ai-news-skill
```

**Antigravity Agent**
```bash
git clone https://github.com/frankzch/ai-news-skill.git ~/.agents/skills/ai-news-skill
```

**Codex**
```bash
git clone https://github.com/frankzch/ai-news-skill.git ~/.codex/skills/ai-news-skill
```

### 🎛️ Modifying Settings & Filters
No complex configuration needed — just control your Agent through natural conversation. For example:
- "Show me AI videos released by top KOLs in the past 5 days"
- "Give me today's AI news, but exclude TechReview"
- "Fetch Reddit AI discussions from the past 3 days"

The agent automatically understands your intent and translates it into precise filtering conditions. You can control:

- 📂 **Filter by Category** — Include or exclude content types: News, Open Source, Social Discussions, KOL Insights
- 📡 **Filter by Source** — Include or exclude specific platforms (e.g., Reddit, TechCrunch, etc.)
- ⏰ **Time Range** — Customize how far back to fetch (default: past 24 hours)
- 🔢 **Result Limit** — Control the number of items returned (per-category cap depends on tier: guest top 5 / logged-in top 10 / member all)
- 📝 **Summary Display** — Choose whether to show short summaries, long summaries
- 🔗 **Link Display** — Choose whether to show original article links
- 🌐 **Language** — Choose output language (English or Chinese). Defaults to your system language; falls back to English if neither
- 💾 **Output Format** — Results can be printed directly to the terminal or exported as a JSON file

### 🔧 AgentSkills Compatible
This skill follows the [AgentSkills](https://agentskills.io/) open standard. The `SKILL.md` file contains all metadata and instructions needed for any compatible agent to discover, activate, and execute this skill automatically.

### ❓ FAQ

**How do I get the latest AI news inside Claude Code or Codex?**
Install this skill (one `git clone`), then ask in plain language, e.g. "AI news from the past 24 hours, news only, long summaries". The agent turns it into category / source / time-range filters.

**Can I follow AI builders and KOLs on X without scrolling X?**
Yes. High-engagement posts from ~310 AI builders are collected and summarized; ask for "KOL insights from the past 3 days".

**Is it free?**
Yes. Without an account you get the top 5 items in each category; a free InBrief.info account raises it to the top 10; members get everything.

**Can I get the digest in Chinese?**
Yes. Every item is summarized in both English and Chinese; ask for Chinese, or it follows your system language.

**Can I self-host the pipeline instead?**
Yes — the collection and curation engine is open source: [ai-news-brief](https://github.com/frankzch/ai-news-brief) (MIT).

### 🤝 Contributing
We welcome contributions! Please see [CONTRIBUTING.md](./CONTRIBUTING.md) for guidelines.

### 📄 License
This project is licensed under the [MIT License](./LICENSE).
````
