Update wiki documentation for current repo state

- Rewrite Flow Deploy section: ephemeral CI runtimes, test YAML config,
  provision/teardown lifecycle, Check Run blocking
- Add Postgres CDC Demo flow page
- Update Flows index and sidebar with data-generator bucket

.... Generated with [Cortex Code](https://docs.snowflake.com/en/user-guide/cortex-code/cortex-code)

Co-Authored-By: Cortex Code <noreply@snowflake.com>

Author: sfc-gh-pvillard

wiki/Flows--Postgres-CDC-Demo.md

@@ -0,0 +1,50 @@
+# Postgres CDC Demo — Data Generator
+
+**Bucket:** `data-generator`
+**File:** [`flows/data-generator/postgres-cdc-demo.json`](../flows/data-generator/postgres-cdc-demo.json)
+
+Simulates random data generation with INSERTs, UPDATEs, and DELETEs across three tables (`customers`, `orders`, `order_items`). Designed to be used in combination with the CDC PostgreSQL Connector to demonstrate change data capture pipelines.
+
+---
+
+## Purpose
+
+This flow provides a continuous stream of realistic database mutations against a PostgreSQL instance. It automatically creates the required schema, tables, and publication on first run — no manual SQL setup is needed.
+
+## Components
+
+| Component | Type | Description |
+|-----------|------|-------------|
+| ExecuteScript (init) | Processor | Creates schema, tables, and publication on first run |
+| GenerateFlowFile | Processor | Triggers periodic data generation |
+| ExecuteSQL | Processor | Runs randomised INSERT/UPDATE/DELETE statements |
+
+## Required NARs
+
+- `org.apache.nifi:nifi-standard-nar:2.8.0` — included with standard NiFi installations
+- PostgreSQL JDBC driver (`postgresql-42.7.10.jar`) — uploaded as a Parameter Context Asset
+
+## Parameters
+
+| Parameter | Description |
+|-----------|-------------|
+| `Database Connection URL` | JDBC URL to the Postgres instance |
+| `Database Name` | Postgres database name |
+| `Database User` | Postgres username |
+| `Database Password` | Postgres password (sensitive — use parameter provider) |
+| `Schema Name` | Schema for the generated tables |
+| `Database Driver` | JDBC driver asset (bound to the uploaded JAR) |
+
+## Configuration
+
+Deploy via the Environment CD pipeline using an `environments/<env>/config.yaml` entry, or test in CI using the test YAML at `flows/data-generator/tests/test_postgres_cdc_demo.yaml`.
+
+For secrets (`Database Connection URL`, `Database Password`), use the auto-provisioned Snowflake Parameter Provider with `#{PARAM_NAME}` references.
+
+## Expected Behaviour
+
+Once started, the flow continuously generates random INSERT, UPDATE, and DELETE operations against the three tables. The PostgreSQL publication enables downstream CDC connectors to capture these changes in real time.
+
+## Validation Tests
+
+The test file at [`flows/data-generator/tests/test_postgres_cdc_demo.py`](../flows/data-generator/tests/test_postgres_cdc_demo.py) validates the flow deploys correctly and processes data against a live PostgreSQL instance provisioned via the ephemeral CI runtime.

wiki/Flows.md

@@ -14,6 +14,12 @@ See [How to Use This Repo](How-to-Use-This-Repo) for instructions on importing o
 |------|-------------|
 | [Hello World](Flows--Hello-World) | Minimal example demonstrating the NiFi Hub flow structure. Generates a FlowFile and logs its attributes. |
 
+### Data Generator
+
+| Flow | Description |
+|------|-------------|
+| [Postgres CDC Demo](Flows--Postgres-CDC-Demo) | Simulates random data generation (INSERTs, UPDATEs, DELETEs) across multiple tables for use with the CDC PostgreSQL Connector. |
+
 ---
 
 ## Flow Structure

wiki/Introduction-and-Concepts--CD.md

@@ -1,6 +1,6 @@
 # CD Pipeline
 
-NiFi Hub has two CD mechanisms: **Flow Deploy** for testing flows against a live Snowflake runtime during PR review, and **Environment CD** for managing Openflow infrastructure declaratively as code.
+NiFi Hub has two CD mechanisms: **Flow Deploy** for testing flows against ephemeral Snowflake runtimes during PR review, and **Environment CD** for managing Openflow infrastructure declaratively as code.
 
 ---
 
@@ -9,22 +9,47 @@ NiFi Hub has two CD mechanisms: **Flow Deploy** for testing flows against a live
 **Workflow:** `flow-deploy.yml`
 **Trigger:** A maintainer with admin or maintain permission comments `deploy this flow` on a PR
 
-This workflow lets maintainers test a flow against a real Snowflake Openflow runtime before merging. It is intended as a validation step during PR review, not for production deployment.
+This workflow lets maintainers test a flow against a real Snowflake Openflow runtime before merging. It provisions an **ephemeral runtime**, deploys the flow, runs tests, and tears everything down automatically.
 
 ### What Happens
 
 1. The workflow identifies flow JSON files changed in the PR
-2. **Builds all extension bundles** and uploads the resulting NARs to the target runtime
-3. **Deploys each changed flow** as a process group on the runtime, using the runtime's REST API
-4. **Runs the flow's validation tests** (`flows/<bucket>/tests/test_<flow-name>.py`) against the deployed process group
-5. **Cleans up** the deployed process group and uploaded NARs (unless the comment includes "do not clean")
-6. **Posts a comment** on the PR with deployment details, processor/controller service summary, and per-test results
-
-### Configuration
-
-The target runtime is configured via a GitHub Environment named `snowflake-runtime-ci`, which provides:
-- `SNOWFLAKE_RUNTIME_URL` — the Openflow runtime endpoint
-- `SNOWFLAKE_RUNTIME_PAT` — a PAT with permission to deploy to the runtime
+2. For each changed flow that has a test YAML (`flows/<bucket>/tests/test_<flow-name>.yaml`):
+   - **Provisions an ephemeral runtime** named `CI_<FLOW>_<PR>_<RUN_ID>` with the configuration from the test YAML (node type, network rules, registries, etc.)
+   - **Uploads custom NARs** from GitHub Releases if specified in the test YAML's `nars` field
+   - **Deploys the flow** from the PR branch as a process group on the runtime
+   - **Applies parameters and assets** from the test YAML's `flow` section
+   - **Fetches the auto-provisioned Snowflake Parameter Provider** to inject secrets from Snowflake
+   - **Waits** for the flow to process data (60 seconds)
+   - **Runs validation tests** (`flows/<bucket>/tests/test_<flow-name>.py`) against the deployed process group
+   - **Tears down** the ephemeral runtime (unless the comment includes "do not clean")
+3. **Posts a comment** on the PR with per-test results and failure details
+4. **Creates a Check Run** that blocks the PR merge if tests fail
+
+### Test YAML Configuration
+
+Each flow that needs CI testing has a `flows/<bucket>/tests/test_<flow-name>.yaml` file following the [CI runtime schema](../scripts/ci/ci-runtime-schema.json). Key fields:
+
+| Field | Description |
+|-------|-------------|
+| `github_environment` | GitHub Environment holding Snowflake secrets |
+| `deployment` | Existing Openflow deployment to create the ephemeral runtime in |
+| `database` / `schema` | Where the runtime is scoped |
+| `node_type` | Runtime size (`SMALL`, `MEDIUM`, `LARGE`) |
+| `execute_as_role` | Snowflake role for runtime data access |
+| `network_rules` | Egress rules needed by the flow under test |
+| `flow_registries` | Git registry clients to configure |
+| `nars` | Custom NARs to upload (URLs or `${GITHUB_RELEASES}/` references) |
+| `sensitive_param_pattern` | Regex for classifying auto-provisioned provider parameters as sensitive |
+| `flow.parameters` | Parameter values to apply after deployment |
+| `flow.assets` | Files to download and bind as parameter context assets |
+
+### GitHub Environment Secrets
+
+The GitHub Environment referenced by `github_environment` must provide:
+- `SNOWFLAKE_PAT` — Snowflake PAT for provisioning the ephemeral runtime
+- `NIFI_RUNTIME_PAT` — PAT for NiFi REST API operations
+- `NIFIHUB_REGISTRY_PAT` — GitHub PAT for the Flow Registry Client
 
 ---
 

wiki/_Sidebar.md

@@ -24,3 +24,4 @@
 
 **[Flows](Flows)**
 - [Hello World](Flows--Hello-World)
+- [Postgres CDC Demo](Flows--Postgres-CDC-Demo)