Automate a weekly RSS newsletter draft
One person uses OpenClaw to gather feed items, rank them, draft intros, and wait for approval before sending the newsletter.
- What they did
- They connected five markdown skills with cron jobs and a heartbeat check. Every six hours, the agent collects roughly 120 RSS sources, plus Reddit and Hacker News, into one inbox. On Wednesday night it rates the items by relevance, freshness, source credibility, engagement, and repeat penalties. On Thursday it writes short openings in the editor’s style and sends a Telegram preview for review.
- How it went
- The workflow is claimed to save about 18 minutes per issue, compared with around four hours by hand. The first few sends should go out as drafts in Buttondown so formatting can be checked, and the intro text can still miss subtle context.
- Tips
- Keep a person in the loop for final review, especially on the opening copy. Use drafts at first, and don’t let the agent send anything until you reply with approval.
Try it yourself with OpenClaw
You are my newsletter assistant. Watch these RSS feeds: [feed URLs], plus [optional Reddit/Hacker News sources]. Every [time interval], collect new items into an inbox. Once a week, rank them for [topic/audience] using relevance, freshness, source trust, engagement, and duplicate penalties. Draft a newsletter with a brief intro in the style of [editor voice] and [one line / short blurb] for each selected item. Send me a preview in [Telegram or other channel]. Pause and wait for my reply of "approve" before you send, publish, or submit anything to [newsletter tool/list]. If I ask for changes, revise and resend the preview first.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
1 more report of this
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Use an agent to draft and publish SEO blog posts
A writer set up OpenClaw to mimic their style, spot search opportunities, and prepare posts for manual approval before publishing.
Edit a video in chat and export the finished file
A developer made an OpenClaw skill that turns chat requests into backend video edits and returns the rendered file.
Make a daily audio briefing from your newsletters
They had an agent read selected AI newsletters overnight and turn the highlights into a podcast episode ready for the morning.
Make an Ableton Live track from chat prompts
A developer used Codex with an MCP server to build and tweak a full Ableton Live song by conversation.
Render a video, then post it in Slack manually
When Slack upload support failed, an agent revised the video, re-rendered it, and finished the post with Computer Use.
Turn one idea into a finished video
Grok’s Agent Mode took a concept from brainstorming through writing, images, edits, and final video in one canvas.