Local-first AI memory system that stores conversation history verbatim and retrieves it with semantic search, exposed as an MCP server and CLI.
Thysrael/Horizon
A self-hosted Python pipeline that collects news from many sources, scores and deduplicates it, and writes a daily briefing in English and Chinese.
Open-source alternative to:Feedly
- Stars
- 10,793
- Forks
- 1,636
- Open issues
- 1
- Last commit
- yesterday
What is it?
Horizon collects items from sources such as RSS, Hacker News, Reddit, Telegram, X, GitHub, and financial news, then filters and deduplicates them before producing a briefing. Each source is routed to a processing profile that defines its own scoring criteria and output format, so a technical blog post and a news story can be treated differently.
Selected items are enriched with web-researched background and community discussion where useful, then grouped by profile into a bilingual Markdown briefing. The output can be published to GitHub Pages, sent by email, or pushed to Feishu/Lark, DingTalk, Slack, Discord, or a custom webhook. Category limits keep one busy topic from crowding out the rest.
The project is run through a CLI, a Docker setup, or an interactive wizard that generates the configuration. An MCP server lets an AI assistant call its pipeline stages directly.
Key features
- 01Collects from RSS, Hacker News, Reddit, Telegram, X, GitHub, OpenBB and other sources
- 02Scores items with per-profile rubrics and thresholds, and merges duplicate stories
- 03Generates English and Chinese Markdown briefings
- 04Delivers via GitHub Pages, email, WeChat, and webhooks including Feishu and Slack
- 05Supports an MCP server so AI assistants can run pipeline stages
Use cases
- Keeping up with tech and AI news without scanning many feeds by hand
- Building a personal or team daily digest published as a static site
- Monitoring financial or industry news and pushing it to a chat group
- Collecting material for content creation with profiles tuned for creators
Quick start
- 1
Clone the repository
git clone https://github.com/Thysrael/Horizon.git - 2
Enter the project folder
cd Horizon - 3
Install dependencies with uv
uv sync - 4
Run the interactive setup wizard to generate config
uv run horizon-wizard
Things to know
- Requires an AI provider API key (e.g. OpenAI) to score and summarize news.
- The optional OpenBB source may need binary-only installs on some machines.
Topics
FAQ
- What is Horizon?
- A self-hosted Python pipeline that collects news from many sources, scores and deduplicates it, and writes a daily briefing in English and Chinese. Horizon collects items from sources such as RSS, Hacker News, Reddit, Telegram, X, GitHub, and financial news, then filters and deduplicates them before producing a briefing. Each source is routed to a processing profile that defines its own scoring criteria and output format, so a technical blog post and a news story can be treated differently.
- Is Horizon free to use? What license does it have?
- Horizon is open source under the MIT license. Check the license terms before using it in commercial products.
- Is Horizon still maintained?
- Actively maintained. The last commit to Horizon was on Oct 19, 2026.
- What can I use Horizon for?
- Keeping up with tech and AI news without scanning many feeds by hand Building a personal or team daily digest published as a static site Monitoring financial or industry news and pushing it to a chat group Collecting material for content creation with profiles tuned for creators
- How do I install Horizon?
- Clone the repository: git clone https://github.com/Thysrael/Horizon.git Enter the project folder: cd Horizon Install dependencies with uv: uv sync Run the interactive setup wizard to generate config: uv run horizon-wizard
- Where can I get Horizon?
- The source code and installation instructions are on GitHub: https://github.com/Thysrael/Horizon
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Agent memory system that stores, recalls and reflects on memories so AI agents learn over time instead of only replaying chat history.
TypeScript and Python SDK giving AI agents 1000+ pre-authenticated app toolkits, per-user sessions, and a sandbox to take actions.
A minimal Python library for building AI agents that write and run their actions as Python code.
A Python framework for building MCP servers, clients, and interactive apps that connect LLMs to tools and data.
Official Python SDK for building Model Context Protocol servers and clients that connect LLM applications to tools, data and prompts.