---
name: gousto
description: Search and browse 9,000+ Gousto recipes. Get full ingredients and step-by-step cooking instructions via official API.
homepage: https://github.com/dhruvkelawala/gousto-agent-skill
metadata: {"openclaw":{"emoji":"🍳","requires":{"bins":["curl","jq"]}}}
---
# Gousto Recipe Skill
Search and browse 9,000+ Gousto recipes from the command line.
## Quick Start
```bash
# First time: build the cache (~3 min)
./scripts/update-cache.sh
# Search recipes
./scripts/search.sh chicken
./scripts/search.sh "beef curry"
# Get full recipe with ingredients & steps
./scripts/recipe.sh honey-soy-chicken-with-noodles
```
## Scripts
| Script | Purpose |
|--------|---------|
| `search.sh <query>` | Search recipes by title (uses local cache) |
| `recipe.sh <slug>` | Get full recipe details with ingredients and cooking steps |
| `update-cache.sh` | Rebuild local cache from Gousto API (~3 min) |
## API Details
**Official Gousto API** (recipe listing):
```
https://production-api.gousto.co.uk/cmsreadbroker/v1/recipes?limit=50&offset=0
```
- Returns metadata: title, rating, prep_time, url
- Paginate with `offset` parameter (NOT `skip` — that's broken!)
- ~9,300 recipes total
**Official Gousto API** (single recipe):
```
https://production-api.gousto.co.uk/cmsreadbroker/v1/recipe/{slug}
```
- Full recipe with ingredients, cooking steps, nutritional info
- HTML in steps is stripped to plain text by the script
## Cache Format
`data/recipes.json` — array of objects:
```json
{
"title": "Chicken Tikka Masala",
"slug": "chicken-tikka-masala",
"rating": 4.8,
"rating_count": 12543,
"prep_time": 35,
"uid": "blt123..."
}
```
## Notes
- Cache is gitignored — run `update-cache.sh` after cloning
- Search is instant (local jq filter)
- Recipe fetch requires network (vfjr.dev proxy)Help answer questions about Catholicism accurately
Analyze budget vs actual
Push decisions to Arbiter Zebu for async human review.
Create, validate, and publish Agent Skills following