RSSidian

RSSidian is an MCP server that connects RSS subscriptions to Obsidian for automated content ingestion, semantic indexing, and AI-driven summarization. Designed for researchers, knowledge workers, and personal knowledge management enthusiasts, it bridges feed aggregators with local markdown note archives. RSSidian imports feed collections via OPML or direct command-line commands, retrieves recent articles, and stores text along with Annoy vector embeddings in a local SQLite database. By integrating OpenRouter, the server evaluates content quality across tiered rating systems, scores relevance, and generates structured digests directly into an Obsidian vault using custom markdown templates. Its Model Context Protocol service allows AI assistants like Claude to query ingested articles, run natural language searches across historical feed data, inspect feed performance metrics, and retrieve full article contents on demand. Users can filter noise by muting feeds or enforcing minimum quality thresholds before export, ensuring that only high-signal articles populate their personal knowledge graph.

Category: Communication & Messaging

Tags: content-feed, obsidian, rss, summarization

Visit RSSidian

How to install and configure RSSidian

1. Clone the repository and navigate to the project directory: bash git clone https://github.com/pedramamini/rssidian.git cd rssidian 2. Create a virtual environment and install the package using pip or uv: bash python -m venv .venv source .venv/bin/activate pip install hatch pip install -e . 3. Initialize the configuration file at ~/.config/rssidian/config.toml: bash rssidian init Edit this file to specify your Obsidian vault path and OpenRouter API key. 4. For Claude Desktop, configure the server in STDIO mode within your Claude settings file: json { "mcpServers": { "rssidian": { "command": "/path/to/rssidian/.venv/bin/rssidian", "args": ["mcp", "--stdio"] } } } Alternatively, run rssidian mcp --port 8080 to operate the server over HTTP.

What you can do with RSSidian

• Querying local RSS article archives using natural language semantic search inside Claude Desktop to find relevant research without exact keyword matches. - Generating automated daily or weekly Obsidian digest notes that summarize top-ranked articles and categorize content by quality rating tiers. - Filtering low-value content from high-volume newsfeeds by applying automated AI quality scoring and minimum quality thresholds before saving. - Reviewing feed analytics and subscription statistics directly from an AI agent to decide which feeds to mute or unmute. - Ingesting OPML feed subscriptions into a local SQLite and Annoy vector store for offline processing and semantic knowledge retrieval.

Key facts

• https://github.com/pedramamini/RSSidian
• Communication & Messaging, Notes & Knowledge Management
• content-feed, obsidian, rss, summarization

Part of MCP Servers

How do I install RSSidian?

Clone the repository from GitHub, create a virtual environment, and install dependencies using hatch or uv in editable mode. Next, run the init command to generate your configuration file, then configure your Obsidian vault path and OpenRouter API credentials before starting the MCP service.

What can RSSidian do?

RSSidian imports RSS subscriptions, indexes articles in an SQLite database with Annoy vector embeddings, and generates automated digests for Obsidian. Through OpenRouter, it summarizes articles, scores content quality, and provides an MCP interface for semantic searching and feed management.

Which MCP clients work with RSSidian?

RSSidian works with Claude Desktop and any client that supports the Model Context Protocol. It can be executed directly as a subprocess using STDIO mode or hosted locally as a RESTful HTTP service on a designated network port.

How does RSSidian filter low-quality articles?

RSSidian evaluates incoming articles using OpenRouter models, assigning an S, A, B, C, or D quality tier and a numerical value score. You can define a minimum tier threshold in the configuration file to automatically omit low-scoring content from digest generation.

How does semantic search work in RSSidian?

RSSidian creates vector embeddings for ingested content and maintains an Annoy index. This allows the MCP server to execute natural language queries that match the contextual meaning of topics rather than relying strictly on exact keyword matches.