CLI

CLI for one-off scrapes, scripting, and CI pipelines. Outputs JSON or NDJSON, so it pipes straight into jq or a warehouse loader.

Get an API key

Quick start

1. Install

# macOS, Linux and Windows
npm install -g chocodata-cli

# or run it without installing
npx chocodata-cli amazon product --query 0143127748

2. Authenticate

# Stores the key in ~/.chocodata/config.json
chocodata login asa_live_YOUR_KEY

# Or set CHOCODATA_API_KEY in your env
export CHOCODATA_API_KEY=asa_live_...

3. Scrape your first product

chocodata amazon product --query 0143127748

# Compact, one JSON object per line
chocodata amazon product --query 0143127748 --ndjson

# A different marketplace
chocodata amazon product --query 0143127748 --domain de

How-to

Run a search and pipe to jq

chocodata bing search --q "wireless earbuds" --count 10 --ndjson | jq -r '.organic_results[].url'

Bulk-scrape IDs from a file

# asins.txt: one ASIN per line
cat asins.txt | chocodata amazon product --stdin-param query --ndjson > products.ndjson

# Straight into jq
cat asins.txt | chocodata amazon product --stdin-param query --ndjson   | jq -r '[.asin, .title, .price] | @csv'

CI pipeline (GitHub Actions)

# .github/workflows/scrape.yml
name: Daily price scrape
on:
  schedule:
    - cron: "0 6 * * *"

jobs:
  scrape:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - run: npm install -g chocodata-cli
      - run: cat asins.txt | chocodata amazon product --stdin-param query --ndjson > prices-$(date +%F).ndjson
        env:
          CHOCODATA_API_KEY: ${{ secrets.CHOCODATA_API_KEY }}
      - uses: actions/upload-artifact@v4
        with:
          path: prices-*.ndjson

Available commands

chocodata <target> <resource> [--param value]   # any of 499 endpoints
chocodata login <api-key>                       # store a key
chocodata whoami                                # which key is in use, masked
chocodata --help                                # usage and examples

# flags
--ndjson                 one compact JSON object per line
--stdin-param <name>     read values from stdin into that parameter
--debug                  log every HTTP call to stderr

Why use the CLI?

One command, any endpoint

Every one of the 499 endpoints is reachable as chocodata <target> <resource>, with that endpoint's own parameters as flags.

Pipe-friendly

Pretty JSON by default, or --ndjson for one object per line. Errors go to stderr, so a partial failure still leaves a clean stream.

Built for cron and CI

Reads CHOCODATA_API_KEY from env and exits non-zero on failure, so CI runners detect it.

Batch from a file

--stdin-param reads values one per line and writes each result as it completes, so you can pipe into a database or queue without buffering the whole run.