RSS AI Digest

Overview

Use this skill to turn RSS/Atom feeds and OPML files into high-signal AI and technical reading digests. Keep the workflow portable: use explicit input/output files, produce Markdown or JSON, and avoid runtime-specific assumptions.

Workflow Selection

• For daily news, key information, quick reading, or "今日/本周重点资讯", prefer subscription-research rss archive-digest when a local research workspace archive exists; otherwise run subscription-research rss digest and follow references/digest-report.md.
• For source cleanup, failed-feed review, source quality, or registry maintenance, use rss-source-curator instead of including source-maintenance details in the digest.
• For deep research, evidence briefs, research memos, long-form synthesis, or multi-step investigation, use subscription-research-agent.
• For keyword, author, project, or topic monitoring, run subscription-research rss check-new.
• For OPML import, run subscription-research rss import-opml, then use the resulting registry for digest or monitoring.
• For scheduled checks, read references/automation.md and provide a platform-neutral recipe.

Runtime Command

Examples use subscription-research as a compatibility command. Before running commands, resolve the runtime command: