Integration
ShopifySkills
a local OpenClaw plugin, enable product-research, make sure the FastAPI backend is running, and use the included product-research-flow skill.
Install
pip install -r
Configuration Example
{
"plugins": {
"entries": {
"product-research": {
"enabled": true,
"config": {
"backendBaseUrl": "http://127.0.0.1:8000"
}
}
}
}
}
README
# Shopify Skillsets Product Research Pipeline
Research-only product pipeline for an OpenClaw instance. This repo intentionally implements only:
1. `find_demand_products`
2. `match_supplier_products`
3. `score_products`
There is no Telegram, Shopify, approval, or publishing logic in this phase.
## Architecture
- `skills/product-research-flow/SKILL.md` gives OpenClaw lightweight workflow guidance.
- `index.ts` registers three thin OpenClaw tools that POST to the backend.
- `backend/app` contains the FastAPI backend and all business logic.
- Postgres stores research jobs, candidates, supplier matches, and scores.
## Backend Setup
```bash
cd backend
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
```
Set `DATABASE_URL` in `backend/.env`, then run migrations:
```bash
alembic upgrade head
uvicorn app.main:app --reload --port 8000
```
Health check:
```bash
curl http://127.0.0.1:8000/health
```
## OpenClaw Setup
Install or link this repo as a local native plugin:
```bash
openclaw plugins install -l /Users/xdpanda/Documents/Development/ShopifySkillsets
openclaw plugins enable product-research
openclaw gateway restart
```
Configure the plugin if your backend is not on the default URL:
```json
{
"plugins": {
"entries": {
"product-research": {
"enabled": true,
"config": {
"backendBaseUrl": "http://127.0.0.1:8000"
}
}
}
}
}
```
OpenClaw loads workspace skills from `skills/`; the included `product-research-flow` skill tells the agent to call the three tools in order and summarize only scored products.
## Tests
```bash
cd backend
pytest
```
The integration test uses SQLite in memory for speed, while production configuration targets Postgres.
integration
Comments
Sign in to leave a comment