Rss Agent Viewer
Run an agent-friendly RSS/Atom CLI to discover feeds, subscribe, search articles, and import OPML from the terminal.
Overview
RSS Agent Viewer is an agent skill for the Idea phase that gives agents a CLI to discover, subscribe, search, and read RSS/Atom feeds from the terminal.
Install
npx skills add https://github.com/brooksy4503/rss-agent-viewer --skill rss-agent-viewerWhat is this skill?
- Init config and local database with rss-viewer init
- Discover feeds for any URL plus discover-search query mode
- Subscribe with add/remove, read all or one feed, optional slow-feed timeout
- Full command set: search, import/export OPML, cache, cleanup
- npx -y rss-agent-viewer quick start without global install
- 15+ command categories in core workflow and common commands
Adoption & trust: 510 installs on skills.sh; 4 GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need structured feed access and OPML workflows from the shell so your coding agent can research sources without opening a separate reader app.
Who is it for?
Solo builders and agents who monitor blogs, changelogs, and indie news via terminal automation and OPML portability.
Skip if: Teams that only need a polished GUI RSS app with no agent integration, or workflows with no need for feed discovery or CLI caching.
When should I use this skill?
Agent needs to manage RSS feeds, read articles, discover feeds, import OPML, or search feeds from the terminal.
What do I get? / Deliverables
You get a initialized local feed DB, subscribed URLs, and readable/searchable articles ready for summarization or competitor monitoring in later journey steps.
- Local feed database after init
- OPML export file
- Terminal article listings from read/search
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Feed discovery and reading sit earliest in the journey when solo builders scan sources before committing to a product direction. Discover subphase covers finding and aggregating external signals—exactly what RSS discovery, site scrape, and discover-search commands support.
Where it fits
Run discover on competitor domains before you pick a niche to build in.
Read and search cached articles to compile a weekly signal brief for validation.
Subscribe to industry feeds and search headlines when planning newsletter or blog topics.
Let your agent pull latest posts via read and search during automated research tasks.
How it compares
Use instead of piping random curl into one-off parsers when you want discovery, OPML, and search in a single agent-oriented CLI.
Common Questions / FAQ
Who is rss-agent-viewer for?
Indie builders and AI agents on Claude Code, Cursor, or Codex who want RSS/Atom management, discovery, and OPML from the terminal without a dedicated reader UI.
When should I use rss-agent-viewer?
Use during Idea discover/research to scan competitor and niche feeds; during Grow content planning to harvest article ideas; and during Build agent-tooling when your agent must read RSS feeds or run discover-search on demand.
Is rss-agent-viewer safe to install?
It is MIT-licensed and typically run via npx; review the Security Audits panel on this Prism page and treat feed URLs and cached content as untrusted network input before automating fetches.
SKILL.md
READMESKILL.md - Rss Agent Viewer
# RSS Agent Viewer CLI RSS/Atom feed viewer with built-in discovery, caching, and search. ## Quick start ```bash npx -y rss-agent-viewer init npx -y rss-agent-viewer discover https://vercel.com npx -y rss-agent-viewer add https://vercel.com/atom npx -y rss-agent-viewer read ``` ## Core workflow ```bash # Initialize config + database rss-viewer init # Discover feeds for a site rss-viewer discover https://example.com # Subscribe to a feed (optional: --timeout for slow feeds) rss-viewer add https://example.com/feed.xml rss-viewer add https://slow-site.com/feed.xml --timeout 30000 # List feeds and read articles rss-viewer feeds rss-viewer read ``` ## Common commands ```bash rss-viewer init rss-viewer add <url> rss-viewer discover <url> rss-viewer feeds rss-viewer remove <url> rss-viewer read [url] rss-viewer search <query> rss-viewer discover-search <query> rss-viewer import <file> rss-viewer export rss-viewer cache <action> rss-viewer cleanup ``` ## Usage patterns ### Discover and subscribe ```bash rss-viewer discover https://example.com rss-viewer add https://example.com/rss.xml ``` ### Read a single feed ```bash rss-viewer read https://example.com/rss.xml ``` ### Read all feeds (fetches fresh by default) ```bash # Fetches fresh articles from all feeds in parallel rss-viewer read # Use cached data only (skip network requests) rss-viewer read --cached # Limit results rss-viewer read --limit 10 # One latest article per feed (feed-diverse results) rss-viewer read --latest-per-feed --limit 10 # Timeouts (for slow feeds or many feeds) rss-viewer read --timeout 15000 --overall-timeout 180000 ``` ### Clean up invalid feeds ```bash # Remove broken and duplicate feeds rss-viewer cleanup # Preview what would be removed rss-viewer cleanup --dry-run # Only remove broken feeds rss-viewer cleanup --broken # Only remove duplicates rss-viewer cleanup --duplicates ``` ### Search across all feeds ```bash # Local database search (enhanced with full-text search) rss-viewer search "open source" # Web search + discovery + add + search in one command rss-viewer discover-search "Rust programming" --auto-add --read # Using Exa API (BYOK) rss-viewer discover-search "AI safety" \ --provider exa \ --max-results 5 \ --auto-add \ --read ``` ### Import OPML ```bash rss-viewer import feeds.opml ``` ### Export feeds ```bash rss-viewer export ``` ## Search Options ### Local Database Search ```bash rss-viewer search "React 19" --limit 10 rss-viewer search "TypeScript" --author "Dan" --since "2024-01-01" ``` ### Web Search with Discovery ```bash # Use agent's built-in search (default) rss-viewer discover-search "micro-frontends" --auto-add --read # Use Exa API (requires EXA_API_KEY) export EXA_API_KEY="your-api-key" rss-viewer discover-search "WebGPU" \ --provider exa \ --max-results 5 \ --category Development \ --auto-add \ --read \ --limit 20 ``` ### Configuration **Environment Variables:** ```bash EXA_API_KEY="your-api-key" # Exa API key (optional) RSS_FEED_TIMEOUT=10000 # Per-feed fetch timeout (ms) RSS_VIEWER_SEARCH_PROVIDER="agent|exa" # Search provider (default: agent) RSS_VIEWER_MAX_WEB_RESULTS=10 # Max web search results RSS_VIEWER_SEARCH_LIMIT=20 # Max article results RSS_VIEWER_BOOST_RECENT=false # Boost recent articles in search EXA_API_URL="https://api.exa.ai/search" # Custom Exa endpoint (optional) ``` **Config File (~/.config/rss-viewer/config.json):** ```json { "webSearchProvider": "exa",