openclaw-skills-newsboat_skill

Overview

This skill integrates Newsboat into the agent workflow to read, manage, and archive RSS/Atom feeds from the command line. It provides commands and patterns to list, add, remove, refresh, and extract article content from Newsboat's local files and cache. The skill focuses on practical, reproducible steps for automating feed maintenance and retrieving readable article text.

How this skill works

The skill inspects the Newsboat configuration (/.newsboat/config), the feed list (/.newsboat/urls), and the SQLite cache (~/.newsboat/cache.db). It uses simple shell commands to list feeds, append or remove feed URLs, trigger a refresh, and run SQL queries against rss_item to pull article content. When needed, it pipes HTML content through pandoc to produce clean plain-text articles suitable for summaries or archives.

When to use it

• You want automated access to all feed URLs and metadata stored by Newsboat.
• You need to script adding or removing feeds across machines or backups.
• You want to refresh feeds programmatically (e.g., cron, CI) and ensure cache consistency.
• You need to extract readable article text for indexing, summarization, or archival purposes.
• You maintain an archive or backup of all feed content and want reproducible retrieval.

Best practices

• Keep Newsboat in your PATH or reference absolute binary paths in automation scripts.
• Store a versioned backup of ~/.newsboat/urls before batch edits (sed -i.bak provides this).
• Use newsboat -x reload for non-interactive feed refreshes to avoid UI dependencies.
• Query cache.db with sqlite3 and convert HTML via pandoc to retain readable layout and strip comments.
• Respect feed source terms; avoid overly frequent refreshes to prevent server load and rate-limiting.

Example use cases

• Append a new feed to the master list with echo and distribute the updated file to other systems.
• Remove deprecated feeds with sed while keeping a timestamped backup for auditability.
• Trigger newsboat -x reload from a scheduled job to keep local cache fresh for downstream processing.
• Extract the latest article via sqlite3 and pandoc, then feed the plain text into a summarization or search index pipeline.
• Archive full feed content by exporting rss_item rows and storing them in a centralized backup store.