Recipe: switch clouds with one line

Time: 1 minute · Audience: anyone with a working local contract

Problem

You've prototyped a data product on platform: local (DuckDB / Parquet) and want to ship it to a real cloud — without rewriting the contract.

Change the binding.platform, swap format + location to match the new platform's vocabulary, and re-run fluid apply. The schema, dq rules, accessPolicy, and agentPolicy stay byte-for-byte identical.

Local → GCP / BigQuery

 exposes:
   - exposeId: bitcoin_prices
     binding:
-      platform: local
-      format: parquet
-      location:
-        path: ./runtime/out/bitcoin_prices.parquet
+      platform: gcp
+      format: bigquery_table
+      location:
+        project: my-project
+        dataset: crypto_data
+        table: bitcoin_prices
+        region: europe-west3

pip install "data-product-forge[gcp]"
gcloud auth application-default login
fluid apply contract.fluid.yaml --yes

Local → AWS / Athena

     binding:
-      platform: local
-      format: parquet
-      location:
-        path: ./runtime/out/bitcoin_prices.parquet
+      platform: aws
+      format: s3_file
+      location:
+        bucket: example-data-lake
+        prefix: crypto/bitcoin_prices/
+        region: eu-west-1

pip install "data-product-forge[aws]"
aws configure
fluid apply contract.fluid.yaml --yes

Local → Snowflake

     binding:
-      platform: local
-      format: parquet
-      location:
-        path: ./runtime/out/bitcoin_prices.parquet
+      platform: snowflake
+      format: snowflake_table
+      location:
+        database: PROD
+        schema: GOLD
+        table: BITCOIN_PRICES

pip install "data-product-forge[snowflake]"
export SNOWFLAKE_ACCOUNT=...
export SNOWFLAKE_USER=...
fluid apply contract.fluid.yaml --yes

What stays the same

fluidVersion, kind, id, name, metadata — untouched.
builds[] — untouched if it's embedded-logic SQL.
exposes[].contract.schema — untouched.
exposes[].contract.dq.rules — untouched.
accessPolicy.grants[] — principal: strings change to match the cloud's IAM format (e.g. group: for GCP, role: for Snowflake), but the structure is the same.
agentPolicy — untouched.