ci: publish github workflow separate version tags from master

.github/workflows/publish.yml

@@ -47,7 +47,7 @@ jobs:
           platforms: ${{ matrix.platform }}
           push: true
           tags: |
-            sjdonado/bit:${{ steps.version.outputs.version }}-${{ contains(matrix.platform, 'amd64') && 'amd64' || 'arm64' }}
             sjdonado/bit:latest-${{ contains(matrix.platform, 'amd64') && 'amd64' || 'arm64' }}
+            ${{ github.event_name == 'release' && format('sjdonado/bit:{0}-{1}', steps.version.outputs.version, contains(matrix.platform, 'amd64') && 'amd64' || 'arm64') || '' }}
           cache-from: type=gha
           cache-to: type=gha,mode=max

README.md

@@ -8,6 +8,8 @@ Lightweight URL shortener service with minimal resource requirements. Average me
 
 Bit is highly performant, achieving over 850 requests per second with an average latency of just 118ms. For detailed benchmark results, see [benchmark](docs/SETUP.md#benchmark).
 
+Images available on [Docker Hub](https://hub.docker.com/r/sjdonado/bit/tags).
+
 ## Quick Start
 ```bash
 docker run -p 4000:4000 -e ADMIN_API_KEY=$(openssl rand -base64 32) sjdonado/bit:latest
@@ -18,13 +20,6 @@ docker run -p 4000:4000 -e ADMIN_API_KEY=$(openssl rand -base64 32) sjdonado/bit
 - 50MB RAM (20MB avg usage)
 - x86_64 or ARM64 architecture
 
-## Available Images
-
-| Tag     | Architecture | Compressed Size | Uncompressed Size |
-|---------|--------------|-----------------|-------------------|
-| `latest`| linux/amd64  | 12.12 MB        | 32.3 MB           |
-| `latest`| linux/arm64  | 12.8 MB         | 32.3 MB           |
-
 ## Documentation
 - [API Reference](docs/API.md)
 - [Advanced Setup](docs/SETUP.md)