Update documentation with fixed image tag

Signed-off-by: Mihai Criveti <[email protected]>

Author: crivetimihai

docs/docs/deployment/compose.md

@@ -18,7 +18,7 @@ export COMPOSE_CMD="docker compose"
 ## 🐳/🦭 Build the images
 
 ```bash
-docker pull ghcr.io/ibm/mcp-context-forge:latest
+docker pull ghcr.io/ibm/mcp-context-forge:0.5.0
 ```
 
 ## 🐳/🦭 Build the images (when doing local development)
@@ -71,7 +71,7 @@ COMPOSE_ENGINE=podman make compose-up   # force Podman
 
 | Make target       | Docker CLI                                    | Podman built-in                              | podman-compose                               |
 | ----------------- | --------------------------------------------- | -------------------------------------------- | -------------------------------------------- |
-| `compose-up`      | `docker compose -f podman-compose.yml up -d`  | `podman compose -f podman-compose.yml up -d` | `podman-compose -f podman-compose.yml up -d` |
+| `compose-up`      | `docker compose -f docker-compose.yml up -d`  | `podman compose -f docker-compose.yml up -d` | `podman-compose -f docker-compose.yml up -d` |
 | `compose-restart` | `docker compose up -d --pull=missing --build` | idem                                         | idem                                         |
 | `compose-logs`    | `docker compose logs -f`                      | `podman compose logs -f`                     | `podman-compose logs -f`                     |
 | `compose-ps`      | `docker compose ps`                           | `podman compose ps`                          | `podman-compose ps`                          |
@@ -96,7 +96,7 @@ curl http://localhost:4444/health    # {"status":"ok"}
 
 ## 🗄 Selecting a database
 
-Uncomment one service block in `podman-compose.yml` and align `DATABASE_URL`:
+Uncomment one service block in `docker-compose.yml` and align `DATABASE_URL`:
 
 | Service block         | Connection string                             |
 | --------------------- | --------------------------------------------- |

docs/docs/deployment/container.md

@@ -18,7 +18,7 @@ docker run -d --name mcpgateway \
   -e AUTH_REQUIRED=true \
   -e DATABASE_URL=sqlite:///./mcp.db \
   --network=host \
-  ghcr.io/ibm/mcp-context-forge:latest
+  ghcr.io/ibm/mcp-context-forge:0.5.0
 
 docker logs mcpgateway
 ```

docs/docs/deployment/google-cloud-run.md

@@ -16,7 +16,7 @@ Google Cloud Run is an ideal platform for MCP Gateway due to its:
 You can deploy the public image directly:
 
 ```text
-ghcr.io/ibm/mcp-context-forge:latest
+ghcr.io/ibm/mcp-context-forge:0.5.0
 ```
 
 ---
@@ -336,7 +336,7 @@ mcpgw=> \dt;
 Use the MCP Gateway container to generate a JWT token:
 
 ```bash
-docker run -it --rm ghcr.io/ibm/mcp-context-forge:latest \
+docker run -it --rm ghcr.io/ibm/mcp-context-forge:0.5.0 \
   python3 -m mcpgateway.utils.create_jwt_token -u admin --secret jwt-secret-key
 ```
 

docs/docs/deployment/local.md

@@ -9,7 +9,7 @@ This guide walks you through running MCP Gateway on your local machine using a v
 The easiest way to start the server in development mode:
 
 ```bash
-make venv install serve
+make venv install-dev serve
 ```
 
 This does the following:
@@ -25,7 +25,8 @@ This does the following:
 If you want auto-reload on code changes:
 
 ```bash
-make run        # or:
+make dev        # hot-reload (Uvicorn) on :8000
+# or:
 ./run.sh --reload --log debug
 ```
 

docs/docs/deployment/minikube.md

@@ -2,7 +2,7 @@
 
 1. **Install Minikube and kubectl** (Docker or Podman driver required).
 2. Start a local cluster with Ingress and DNS addons.
-3. Load the `ghcr.io/ibm/mcp-context-forge:latest` image into Minikube.
+3. Load the `ghcr.io/ibm/mcp-context-forge:0.5.0` image into Minikube.
 4. Apply your Kubernetes manifests.
 5. Access the Gateway at [http://gateway.local](http://gateway.local) or `127.0.0.1:80` via NGINX Ingress.
 
@@ -131,21 +131,21 @@ kubectl get pods -n ingress-nginx
 make minikube-image-load
 ```
 
-This target builds the `ghcr.io/ibm/mcp-context-forge:latest` image and loads it into Minikube.
+This target builds the `ghcr.io/ibm/mcp-context-forge:0.5.0` image and loads it into Minikube.
 
 ### Alternative methods
 
 * **Pre-cache a remote image:**
 
   ```bash
-  minikube cache add ghcr.io/ibm/mcp-context-forge:latest
+  minikube cache add ghcr.io/ibm/mcp-context-forge:0.5.0
   minikube cache reload
   ```
 
 * **Load a local tarball:**
 
   ```bash
-  docker save ghcr.io/ibm/mcp-context-forge:latest | minikube image load -
+  docker save ghcr.io/ibm/mcp-context-forge:0.5.0 | minikube image load -
   ```
 
 ---
@@ -282,7 +282,7 @@ curl http://gateway.local/health
 | ------------------- | ---------------------- | ------------------------------------------------------------ |
 | Pause cluster       | `make minikube-stop`   | `minikube stop -p mcpgw`                                     |
 | Delete cluster      | `make minikube-delete` | `minikube delete -p mcpgw`                                   |
-| Remove cached image | -                      | `minikube cache delete ghcr.io/ibm/mcp-context-forge:latest` |
+| Remove cached image | -                      | `minikube cache delete ghcr.io/ibm/mcp-context-forge:0.5.0` |
 
 ---
 

docs/docs/development/building.md

@@ -6,7 +6,7 @@ Follow these instructions to set up your development environment, build the gate
 
 ## 🧩 Prerequisites
 
-- Python **≥ 3.10**
+- Python **≥ 3.11**
 - `make`
 - (Optional) Docker or Podman for container builds
 
@@ -15,13 +15,13 @@ Follow these instructions to set up your development environment, build the gate
 ## 🔧 One-Liner Setup (Recommended)
 
 ```bash
-make venv install serve
+make venv install-dev serve
 ```
 
 This will:
 
 1. Create a virtual environment in `.venv/`
-2. Install Python dependencies including dev extras
+2. Install Python dependencies (including dev extras)
 3. Run the gateway using Gunicorn
 
 ---
@@ -47,12 +47,13 @@ This installs:
 You can run the gateway with:
 
 ```bash
-make serve         # production-mode Gunicorn (http://localhost:4444)
-make run           # dev-mode Uvicorn (reloads on change)
-./run.sh --reload  # same as 'make run', with CLI flags
+make serve         # production-mode (Gunicorn) on http://localhost:4444
+make dev           # hot-reload (Uvicorn) on http://localhost:8000
+make run           # wrapper over uvicorn; pass --reload to enable auto-reload
+./run.sh --reload  # equivalent of 'make run' with explicit flags
 ```
 
-Use `make run` or `./run.sh` during development for auto-reload.
+Use `make dev` during development for auto-reload on port 8000.
 
 ---
 

docs/docs/development/developer-onboarding.md

@@ -7,7 +7,7 @@
 ## 🛠 Environment Setup
 
 ???+ check "System prerequisites"
-    - [ ] Python ≥ 3.10
+    - [ ] Python ≥ 3.11
     - [ ] Node.js and npm, npx (used for testing with `supergateway` and the HTML/JS Admin UI)
     - [ ] Docker, Docker Compose, and Podman
     - [ ] Make, GitHub CLI (`gh`), `curl`, `jq`, `openssl`

docs/docs/development/documentation.md

@@ -6,7 +6,7 @@ Follow this guide when you need to add or update markdown pages under `docs/` an
 
 ## 🧩 Prerequisites
 
-* **Python ≥ 3.10** (only for the initial virtual env - *not* required if you already have one)
+* **Python ≥ 3.11** (only for the initial virtual env - *not* required if you already have one)
 * `make` (GNU Make 4+)
 * (First-time only) **[`mkdocs-material`](https://squidfunk.github.io/mkdocs-material/)** and plugins are installed automatically by the *docs* `Makefile`.
 * One-time GitHub setup, e.g. [gitconfig setup](./github.md#16-personal-git-configuration-recommended)

docs/docs/faq/index.md

@@ -16,7 +16,7 @@
     OCI image (Docker/Podman) - shares host network so localhost works:
 
     ```bash
-    podman run --network=host -p 4444:4444 ghcr.io/ibm/mcp-context-forge:latest
+    podman run --network=host -p 4444:4444 ghcr.io/ibm/mcp-context-forge:0.5.0
     ```
 
 ???+ example "🗂️ What URLs are available for the admin interface and API docs?"
@@ -108,7 +108,7 @@
     Include a persistent volume with your container or Kubernetes deployment. Ex:
 
     ```bash
-    docker run -v $(pwd)/data:/app ghcr.io/ibm/mcp-context-forge:latest
+    docker run -v $(pwd)/data:/app ghcr.io/ibm/mcp-context-forge:0.5.0
     ```
 
     For production use, we recommend PostgreSQL. A Docker Compose target with PostgreSQL and Redis is provided.
@@ -224,9 +224,9 @@
                    token=os.environ["MCPGATEWAY_BEARER_TOKEN"])
     ```
 
-???+ example "🦾 How do I connect GitHub's mcp-server-git via SuperGateway?"
+???+ example "🦾 How do I connect GitHub's mcp-server-git via Translate Bridge?"
     ```bash
-    npx -y supergateway --stdio "uvx mcp-server-git"
+    python3 -m mcpgateway.translate --stdio "uvx mcp-server-git" --expose-sse --port 9001
     ```
 
 ---

docs/docs/manage/logging-examples.md

@@ -108,7 +108,7 @@ LOG_FOLDER=logs
 # docker-compose.yml
 services:
   mcpgateway:
-    image: ghcr.io/ibm/mcp-context-forge:latest
+    image: ghcr.io/ibm/mcp-context-forge:0.5.0
     environment:
       - LOG_LEVEL=INFO
       # Default: logs to stdout/stderr only (recommended for containers)