Skip to main content

AI-powered RSS reader middleware — MCP server connecting Miniflux/FreshRSS to Ollama for smart feed processing

Project description

📡 RSS AI Middleware

Transform your RSS feeds into actionable intelligence.

License MCP Docker PyPI

RSS AI Middleware empowers information-overloaded researchers and power users to eliminate manual feed-scrolling by transforming their curated RSS readers into an interactive AI intelligence agent for instant search, summarization, and organization.


😫 The Friction vs. ⚡ The Flow

The Old Way (Manual) The AI Way (Middleware)
⏳ Hours spent scrolling through 100+ articles 🚀 Seconds to find the exact update you need
🤯 Information overload and "tab fatigue" 🎯 Precise, AI-generated summaries of key points
🔍 Manual keyword searching across multiple feeds 🤖 Natural language queries across your entire library
📁 Tedious manual tagging and starring 🪄 Auto-organization based on intent and priority

🎯 Designed For

  • Researchers: Who need to track niche academic or industry trends without reading every single post.
  • Developers: Who follow 50+ tech blogs and need to know exactly when a specific library updates.
  • News Junkies: Who want a "Daily Briefing" generated from their own curated sources, not an algorithm.
  • AI Agents: Who need a real-time "eye" on the web to feed current data into their workflows.

🛠 The Intelligence Suite

🔍 Deep Discovery

  • Semantic Search: Find entries based on meaning, not just keywords.
  • Cross-Feed Analysis: Query across Miniflux and FreshRSS simultaneously.
  • Instant Retrieval: Get the full content of any entry without leaving your AI chat.

✍️ Synthesis & Insight

  • Token-by-Token Summaries: Get real-time, streaming summaries via Ollama.
  • Multi-Language Translation: Convert global news into your native language instantly.
  • Key Takeaway Extraction: Turn long-form articles into bulleted action items.

🗂 Smart Management

  • Intent-Based Starring: "Star all articles related to Wayland stability."
  • Bulk Status Updates: Mark entire categories as read based on AI filtering.
  • Category Intelligence: List and organize feeds by topic automatically.

💎 The Unfair Advantage (USP)

Unlike generic AI news apps, RSS AI Middleware puts you in total control:

  • BYOK (Bring Your Own Key): Use OpenAI, Anthropic, or Google Gemini.
  • Local-First Privacy: Full support for Ollama. Your news and your keys never leave your machine.
  • MCP Standard: Plug-and-play compatibility with Claude Desktop, Cursor, and OpenCode.
  • Reader Agnostic: Works seamlessly with both Miniflux and FreshRSS.

⚙️ How it Works

Your RSS Reader $\rightarrow$ RSS AI Middleware (MCP) $\rightarrow$ LLM (Ollama/OpenAI) $\rightarrow$ Your AI Agent

  1. Connect: Link your Miniflux/FreshRSS instance via .env.
  2. Bridge: The middleware exposes your feeds as "Tools" to the LLM.
  3. Interact: Ask your AI: "What are the top 3 updates on AI agents from my feeds today?"
  4. Execute: The AI searches, reads, summarizes, and returns the answer.

🚀 Quick Start

1. Clone & Install

```bash git clone https://github.com/ziuus/rss-ai-middleware.git cd rss-ai-middleware python -m venv venv source venv/bin/activate pip install -r requirements.txt ```

2. Configure

Update your .env with your RSS credentials and Ollama URL.

3. Run

```bash python -m src.server ```


🌟 Community & Trust

"Finally, a way to actually use my RSS feeds instead of just hoarding them. The Ollama integration is a game changer for privacy." — Early Beta User

"The MCP implementation is flawless. I have my news feed integrated directly into my coding agent, and it's a productivity superpower." — Open Source Contributor

Trusted by developers and researchers who value their time and their privacy.


📄 License

MIT © ziuus

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rss_ai_middleware-0.2.0.tar.gz (19.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

rss_ai_middleware-0.2.0-py3-none-any.whl (16.5 kB view details)

Uploaded Python 3

File details

Details for the file rss_ai_middleware-0.2.0.tar.gz.

File metadata

  • Download URL: rss_ai_middleware-0.2.0.tar.gz
  • Upload date:
  • Size: 19.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for rss_ai_middleware-0.2.0.tar.gz
Algorithm Hash digest
SHA256 1bdcf13efa0511c1c20bfdb9f8a5432dea41c8996b99c3c29d345a0ef75cb25a
MD5 83a020f173fe2a2ce0891707e915c1c5
BLAKE2b-256 b838a8a55702946c590a20d04d18b0187da67a5e1847e23abfc832371c3c4cfb

See more details on using hashes here.

File details

Details for the file rss_ai_middleware-0.2.0-py3-none-any.whl.

File metadata

File hashes

Hashes for rss_ai_middleware-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 edb08814de3e4c2916f9f51ada931c4d9be72474c604fa22288cd1a65ec559f9
MD5 958c39c74a86cfd95441fe34483d1632
BLAKE2b-256 c3cbaf7a10c4b1bdd6838bfb3f7c42c5accb9b12408efa466fc3dac12e4223e7

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page