Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Privacy & security: Removed personal names from filenames and code (thibaut.md → user1.md, partner.md → user2.md). Generalized all documentation. No personal data in distribution. Local-only operation, no external API calls or credentials.
What This Skill Does
Suggests quick dinner recipes (≤25 minutes) based on your current kitchen inventory and stored taste profiles for two people. Tracks ingredient usage after cooking and builds shopping lists from consumption patterns. A feedback loop updates suggestions over time based on what you liked or disliked.
Combines inventory tracking, dual-person preference matching, and learned feedback in one place rather than mentally cross-referencing what's in the fridge with who eats what.
When to use it
- Getting a weeknight dinner idea using only ingredients already at home
- Tracking what you used after cooking to keep stock accurate
- Building a grocery list from items running low
- Balancing two people's dietary preferences and dislikes in one suggestion
- Avoiding the same meals repeating across the week
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: What should we make for dinner tonight?
- 1Reads current ingredient stock from inventory/stock.md
- 2Checks both taste profiles in preferences/user1.md and preferences/user2.md
- 3Filters available recipes to those completable in ≤25 minutes with on-hand ingredients
- 4Excludes recipes matching either person's dislikes
- 5Selects a suggestion from the rotation and logs it to history
A specific recipe using available ingredients, with prep steps and a note on which items will need restocking