Your first Chef project.

1. Prepare your packages

Clone the repository and pack both packages. Chef builds during packing; the packages are not yet published to npm.

1git clone https://github.com/EndPx/zeer.git2cd zeer3npm ci4mkdir -p .tmp/chef-packages5npm pack ./packages/prep-client --pack-destination .tmp/chef-packages6npm pack ./packages/chef --pack-destination .tmp/chef-packages

2. Create an independent project

Choose a start height your testnet Prep archive covers. The target directory must not already exist.

1node packages/chef/dist/cli.js init \2  --dir ../my-indexer \3  --project my_indexer \4  --start-height 056cd ../my-indexer7npm install ../zeer/.tmp/chef-packages/zeer-prep-client-0.1.0.tgz \8  ../zeer/.tmp/chef-packages/zeer-chef-0.1.0.tgz9npm run codegen10npm run typecheck

3. Connect your data

Edit the generated, git-ignored .env: set your actual PREP_URL, optional Prep token, and PostgreSQL DATABASE_URL. Use the public testnet endpoint below or your own Prep source. Check its current coverage before indexing; the preview requires no personal API token.

1PREP_URL=https://zeer.endpx.cloud/api/prep2PREP_API_TOKEN=

The starter includes optional PostgreSQL via Docker Compose. Skip this command when using an existing database.

1npm run db:up2npm run migrate3npm run status

4. Index, then query

Set an end height available in Prep. Sync uses an inclusive end; Prep queries use an exclusive end. Start the API, then run the query command in another terminal.

1npm run sync -- --to 10022npm run serve34# In another terminal5npm run query

Use npm run follow for ongoing indexing. Restart resumes your persisted checkpoint. Keep tokens server-side, and use a fresh project namespace when changing schema or handler identity.