Let an agent plan meals and daily ideas from your activity

OpenClaw used a user’s past chats, search activity, and Instagram trail to pick restaurants, share daily suggestions, and draft memory edits for review.

Done with OpenClaw· Intermediate· Mar 2026

What they did
A commenter connected OpenClaw to several personal data sources they had already allowed it to read. From that material, the agent scanned old Instagram posts and stories to choose one restaurant each week, then sent three new ideas each day based on the user’s recent patterns and interests. It also prepared memory changes and waited for the user to check and approve them manually.
How it went
The agent booked Osteria Vibrato, which matched where the user had eaten the previous evening. It also pointed to The God of Small Things after noticing a Kerala trip coming up, and the user liked the book.
Tips
Always leave the final say on reservations and stored memories with the person using the agent. Because the example comes from someone involved with building context-use, treat it as a firsthand product account.

Try it yourself with OpenClaw

Look through the ChatGPT, Google, and Instagram data I’ve allowed this agent to access. Learn my preferences from that history, then each week suggest [1] restaurant in [city] for [party size] people on [preferred day/evening], and every morning send me [3] ideas that fit my habits and interests. If you want to reserve a table or change any saved memory, stop first and ask me to approve it, and show me the exact memory edit before it is saved.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Read the original on news.ycombinator.com ↗

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.