What is ClawFeed
ClawFeed is an open-source AI news summarization tool that helps users aggregate information sources across multiple platforms such as Twitter, RSS, HackerNews, Reddit, GitHub, and more, generating structured summaries through AI-powered intelligent filtering (supporting 4-hour/daily/weekly/monthly frequencies). The tool supports features such as Source Packs for sharing curated information bundles, Mark for in-depth analysis, and intelligent filtering, and can be deployed independently or run as a OpenClawor Zylos skill. ClawFeed helps users shift from passively scrolling through feeds to actively acquiring knowledge, saying goodbye to information overload and efficiently gaining knowledge.
ClawFeed's Main Features
- Multi-Source Content Collection: Automatically scrapes Twitter/X, RSS subscriptions, HackerNews, Reddit, GitHub Trending, and content from any website, gathering scattered information streams from various platforms in one place.
- AI Intelligent Summarization: Distills massive amounts of information into structured reports at four frequencies—4-hour, daily, weekly, and monthly—filtering out noise and retaining the essence, greatly saving reading time.
- Source Packs Sharing: Packages curated information sources into themed collections that can be shared with the community in one click or subscribed to from others' recommendations, quickly building a high-quality information network.
- Mark & Deep Dive: Supports adding marks to content of interest, triggering AI for in-depth analysis, trend correlation, and background exploration, moving from "knowing" to "understanding".
- Personalized Filtering Optimization: Supports Feed quality analysis-based intelligent recommendations to follow or unfollow specific sources, continuously optimizing information intake quality.
- Multi-Format Output: Supports Web dashboard browsing, RSS/JSON Feed subscriptions, and Google OAuth multi-user management, meeting the needs of individual reading or team sharing.
How to Use ClawFeed
- Method 1：One-Click Installation via ClawHub (Simplest), execute the installation command clawhub install clawfeed in the terminal, and the system will automatically complete the deployment.
- Method 2：As an OpenClaw Skill, open the terminal and enter the OpenClaw skills directory, execute cd ~/.openclaw/skills/ to switch paths, then execute git clone https://github.com/kevinho/clawfeed.git to clone the project. OpenClaw will automatically detect SKILL.md and load it. AI agents can generate summaries and provide dashboards through scheduled tasks.
- Method 3：As a Zylos Skill, open the terminal and enter the Zylos skills directory, execute cd ~/.zylos/skills/ to switch paths, then execute git clone https://github.com/kevinho/clawfeed.git to clone the project. After Zylos recognizes it, it can be used as a skill.
- Method 4: Independent Deployment (Most Flexible), open the terminal and execute git clone https://github.com/kevinho/clawfeed.git to download the project, execute cd clawfeed to enter the directory, execute npm install to install dependencies, execute cp .env.example .env to copy the environment variable template, then edit the .env file to configure environment variables, execute npm start to start the service. The service runs at http://127.0.0.1:8767。
ClawFeed's Project Address
- GitHub Repository：https://github.com/kevinho/clawfeed
ClawFeed's Application Scenarios
- Industry Trend Tracking: Investors, media professionals, and analysts need to grasp market changes in real time. Through 4-hour high-frequency summaries, they can quickly capture Twitter influencers' opinions, HackerNews technology trends, and GitHub popular projects, avoiding missing key signals.
- Technical Learning Advancement: Developers subscribe to technical blog RSS, GitHub Trending, and Reddit tech communities, with daily summaries automatically aggregating cutting-edge tools and best practices, reducing the time lost switching between multiple platforms.
- Research Material Organization: Researchers and students follow journals and academic updates in specific fields, use Source Packs to package high-quality sources and share them with teams, and the Mark feature flags important literature for AI in-depth analysis of related research.
- Content Creation Assistance: Self-media creators and newsletter authors aggregate trending topics across multiple platforms, form a topic inspiration library after AI screening, and share selected summaries with subscribed readers through the Feed output feature.