PrivyFeed (iOS) — Support

Private, Local AI RSS Reader for iPhone & iPad

This document covers how to get help, frequently asked questions, and troubleshooting for PrivyFeed on iOS.

About PrivyFeed (iOS)

PrivyFeed is a private, local-first application that brings AI to your RSS feeds. It runs entirely using your own Ollama instance, ensuring your reading habits and data never leave your control.

Frequently Asked Questions

How do I add RSS/Atom feeds?

Tap the Add button or Manage Subscriptions in the app.
Enter the URL of the RSS or Atom feed.
Alternatively, use OPML import to bring in all your existing subscriptions.

How does AI Summarization work?

PrivyFeed connects to a local Large Language Model (LLM) — typically hosted on your network via Ollama. When you ask to summarize an article:

The app prepares the article content.
Sends the text to your local LLM.
Streams the summarized results back to you.

How can I Ask Questions across my feeds?

PrivyFeed supports semantic search. It generates embeddings for your articles (saved safely in your device's SQLite database) and retrieves relevant chunks from your reading history when you ask a question. The local LLM then answers based on those chunks.

Where is my data stored?

All subscriptions, articles, indexes, and chat history are stored locally in the app's sandboxed storage on your iPhone or iPad. We don't have accounts or cloud syncing.

System Requirements

iOS 17.0 or later
Ollama or LM Studio: A running instance on your local network (Mac, Linux, Windows, Raspberry Pi) — ollama.ai / lmstudio.ai

How do I connect to Ollama or LM Studio?

Install and run Ollama or LM Studio on your local computer.
Find your computer's local IP address (e.g., 192.168.1.50).
In PrivyFeed Settings, enter the base URL:
- For Ollama: http://192.168.1.50:11434
- For LM Studio: http://192.168.1.50:1234 (assuming default port)
Ensure both devices are on the same Wi-Fi network.

Troubleshooting Connection Issues

If the app says it cannot connect:

Ollama Bind Address: By default, Ollama might only listen on localhost. You need to set the environment variable: OLLAMA_HOST=0.0.0.0 and restart Ollama to allow network connections.
LM Studio: Open the Local Server tab, ensure "CORS" is enabled, and "Listen on network" is checked.
Firewall: Check if your computer's firewall is blocking port 11434 (Ollama) or 1234 (LM Studio).
Network isolation: Some guest Wi-Fi networks block device-to-device communication. Use your private home Wi-Fi instead.

Contact Us

For support, bug reports, or feature requests:

Feedback & Issues: Submit Feedback or Report a Problem