chore!: remove --env from llama stack run (#3711)

# What does this PR do?
user can simply set env vars in the beginning of the command.`FOO=BAR
llama stack run ...`

## Test Plan
Run
TELEMETRY_SINKS=coneol uv run --with llama-stack llama stack build
--distro=starter --image-type=venv --run




---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with
[ReviewStack](https://reviewstack.dev/llamastack/llama-stack/pull/3711).
* #3714
* __->__ #3711

llama_stack/cli/stack/_build.py

@@ -444,12 +444,24 @@ def _run_stack_build_command_from_build_config(
 
         cprint("Build Successful!", color="green", file=sys.stderr)
         cprint(f"You can find the newly-built distribution here: {run_config_file}", color="blue", file=sys.stderr)
-        cprint(
-            "You can run the new Llama Stack distro via: "
-            + colored(f"llama stack run {run_config_file} --image-type {build_config.image_type}", "blue"),
-            color="green",
-            file=sys.stderr,
-        )
+        if build_config.image_type == LlamaStackImageType.VENV:
+            cprint(
+                "You can run the new Llama Stack distro (after activating "
+                + colored(image_name, "cyan")
+                + ") via: "
+                + colored(f"llama stack run {run_config_file}", "blue"),
+                color="green",
+                file=sys.stderr,
+            )
+        elif build_config.image_type == LlamaStackImageType.CONTAINER:
+            cprint(
+                "You can run the container with: "
+                + colored(
+                    f"docker run -p 8321:8321 -v ~/.llama:/root/.llama localhost/{image_name} --port 8321", "blue"
+                ),
+                color="green",
+                file=sys.stderr,
+            )
         return distro_path
     else:
         return _generate_run_config(build_config, build_dir, image_name)

llama_stack/cli/stack/run.py

@@ -16,7 +16,7 @@ import yaml
 from llama_stack.cli.stack.utils import ImageType
 from llama_stack.cli.subcommand import Subcommand
 from llama_stack.core.datatypes import LoggingConfig, StackRunConfig
-from llama_stack.core.stack import cast_image_name_to_string, replace_env_vars, validate_env_pair
+from llama_stack.core.stack import cast_image_name_to_string, replace_env_vars
 from llama_stack.core.utils.config_resolution import Mode, resolve_config_or_distro
 from llama_stack.log import get_logger
 
@@ -55,18 +55,12 @@ class StackRun(Subcommand):
             "--image-name",
             type=str,
             default=None,
-            help="Name of the image to run. Defaults to the current environment",
-        )
-        self.parser.add_argument(
-            "--env",
-            action="append",
-            help="Environment variables to pass to the server in KEY=VALUE format. Can be specified multiple times.",
-            metavar="KEY=VALUE",
+            help="[DEPRECATED] This flag is no longer supported. Please activate your virtual environment before running.",
         )
         self.parser.add_argument(
             "--image-type",
             type=str,
-            help="Image Type used during the build. This can be only venv.",
+            help="[DEPRECATED] This flag is no longer supported. Please activate your virtual environment before running.",
             choices=[e.value for e in ImageType if e.value != ImageType.CONTAINER.value],
         )
         self.parser.add_argument(
@@ -79,11 +73,18 @@ class StackRun(Subcommand):
         import yaml
 
         from llama_stack.core.configure import parse_and_maybe_upgrade_config
-        from llama_stack.core.utils.exec import formulate_run_args, run_command
+
+        if args.image_type or args.image_name:
+            self.parser.error(
+                "The --image-type and --image-name flags are no longer supported.\n\n"
+                "Please activate your virtual environment manually before running `llama stack run`.\n\n"
+                "For example:\n"
+                "  source /path/to/venv/bin/activate\n"
+                "  llama stack run <config>\n"
+            )
 
         if args.enable_ui:
             self._start_ui_development_server(args.port)
-        image_type, image_name = args.image_type, args.image_name
 
         if args.config:
             try:
@@ -95,10 +96,6 @@ class StackRun(Subcommand):
         else:
             config_file = None
 
-        # Check if config is required based on image type
-        if image_type == ImageType.VENV.value and not config_file:
-            self.parser.error("Config file is required for venv environment")
-
         if config_file:
             logger.info(f"Using run configuration: {config_file}")
 
@@ -113,50 +110,13 @@ class StackRun(Subcommand):
                     os.makedirs(str(config.external_providers_dir), exist_ok=True)
             except AttributeError as e:
                 self.parser.error(f"failed to parse config file '{config_file}':\n {e}")
-        else:
-            config = None
 
-        # If neither image type nor image name is provided, assume the server should be run directly
-        # using the current environment packages.
-        if not image_type and not image_name:
-            logger.info("No image type or image name provided. Assuming environment packages.")
-            self._uvicorn_run(config_file, args)
-        else:
-            run_args = formulate_run_args(image_type, image_name)
-
-            run_args.extend([str(args.port)])
-
-            if config_file:
-                run_args.extend(["--config", str(config_file)])
-
-            if args.env:
-                for env_var in args.env:
-                    if "=" not in env_var:
-                        self.parser.error(f"Environment variable '{env_var}' must be in KEY=VALUE format")
-                        return
-                    key, value = env_var.split("=", 1)  # split on first = only
-                    if not key:
-                        self.parser.error(f"Environment variable '{env_var}' has empty key")
-                        return
-                    run_args.extend(["--env", f"{key}={value}"])
-
-            run_command(run_args)
+        self._uvicorn_run(config_file, args)
 
     def _uvicorn_run(self, config_file: Path | None, args: argparse.Namespace) -> None:
         if not config_file:
             self.parser.error("Config file is required")
 
-        # Set environment variables if provided
-        if args.env:
-            for env_pair in args.env:
-                try:
-                    key, value = validate_env_pair(env_pair)
-                    logger.info(f"Setting environment variable {key} => {value}")
-                    os.environ[key] = value
-                except ValueError as e:
-                    logger.error(f"Error: {str(e)}")
-                    self.parser.error(f"Invalid environment variable format: {env_pair}")
-
         config_file = resolve_config_or_distro(str(config_file), Mode.RUN)
         with open(config_file) as fp:
             config_contents = yaml.safe_load(fp)

llama_stack/core/stack.py

@@ -274,22 +274,6 @@ def cast_image_name_to_string(config_dict: dict[str, Any]) -> dict[str, Any]:
     return config_dict
 
 
-def validate_env_pair(env_pair: str) -> tuple[str, str]:
-    """Validate and split an environment variable key-value pair."""
-    try:
-        key, value = env_pair.split("=", 1)
-        key = key.strip()
-        if not key:
-            raise ValueError(f"Empty key in environment variable pair: {env_pair}")
-        if not all(c.isalnum() or c == "_" for c in key):
-            raise ValueError(f"Key must contain only alphanumeric characters and underscores: {key}")
-        return key, value
-    except ValueError as e:
-        raise ValueError(
-            f"Invalid environment variable format '{env_pair}': {str(e)}. Expected format: KEY=value"
-        ) from e
-
-
 def add_internal_implementations(impls: dict[Api, Any], run_config: StackRunConfig) -> None:
     """Add internal implementations (inspect and providers) to the implementations dictionary.
 

llama_stack/core/start_stack.sh

@@ -25,7 +25,7 @@ error_handler() {
 trap 'error_handler ${LINENO}' ERR
 
 if [ $# -lt 3 ]; then
-  echo "Usage: $0 <env_type> <env_path_or_name> <port> [--config <yaml_config>] [--env KEY=VALUE]..."
+  echo "Usage: $0 <env_type> <env_path_or_name> <port> [--config <yaml_config>]"
   exit 1
 fi
 
@@ -43,7 +43,6 @@ SCRIPT_DIR=$(dirname "$(readlink -f "$0")")
 
 # Initialize variables
 yaml_config=""
-env_vars=""
 other_args=""
 
 # Process remaining arguments
@@ -58,15 +57,6 @@ while [[ $# -gt 0 ]]; do
         exit 1
       fi
       ;;
-    --env)
-      if [[ -n "$2" ]]; then
-        env_vars="$env_vars --env $2"
-        shift 2
-      else
-        echo -e "${RED}Error: --env requires a KEY=VALUE argument${NC}" >&2
-        exit 1
-      fi
-      ;;
     *)
       other_args="$other_args $1"
       shift
@@ -119,7 +109,6 @@ if [[ "$env_type" == "venv" ]]; then
     llama stack run \
     $yaml_config_arg \
     --port "$port" \
-    $env_vars \
     $other_args
 elif [[ "$env_type" == "container" ]]; then
     echo -e "${RED}Warning: Llama Stack no longer supports running Containers via the 'llama stack run' command.${NC}"

llama_stack/distributions/dell/doc_template.md

@@ -117,11 +117,11 @@ docker run -it \
   # NOTE: mount the llama-stack directory if testing local changes else not needed
   -v $HOME/git/llama-stack:/app/llama-stack-source \
   # localhost/distribution-dell:dev if building / testing locally
+  -e INFERENCE_MODEL=$INFERENCE_MODEL \
+  -e DEH_URL=$DEH_URL \
+  -e CHROMA_URL=$CHROMA_URL \
   llamastack/distribution-{{ name }}\
-  --port $LLAMA_STACK_PORT  \
-  --env INFERENCE_MODEL=$INFERENCE_MODEL \
-  --env DEH_URL=$DEH_URL \
-  --env CHROMA_URL=$CHROMA_URL
+  --port $LLAMA_STACK_PORT
 
 ```
 
@@ -142,14 +142,14 @@ docker run \
   -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
   -v $HOME/.llama:/root/.llama \
   -v ./llama_stack/distributions/tgi/run-with-safety.yaml:/root/my-run.yaml \
+  -e INFERENCE_MODEL=$INFERENCE_MODEL \
+  -e DEH_URL=$DEH_URL \
+  -e SAFETY_MODEL=$SAFETY_MODEL \
+  -e DEH_SAFETY_URL=$DEH_SAFETY_URL \
+  -e CHROMA_URL=$CHROMA_URL \
   llamastack/distribution-{{ name }} \
   --config /root/my-run.yaml \
-  --port $LLAMA_STACK_PORT \
-  --env INFERENCE_MODEL=$INFERENCE_MODEL \
-  --env DEH_URL=$DEH_URL \
-  --env SAFETY_MODEL=$SAFETY_MODEL \
-  --env DEH_SAFETY_URL=$DEH_SAFETY_URL \
-  --env CHROMA_URL=$CHROMA_URL
+  --port $LLAMA_STACK_PORT
 ```
 
 ### Via Conda
@@ -158,21 +158,21 @@ Make sure you have done `pip install llama-stack` and have the Llama Stack CLI a
 
 ```bash
 llama stack build --distro {{ name }} --image-type conda
-llama stack run {{ name }}
-  --port $LLAMA_STACK_PORT \
-  --env INFERENCE_MODEL=$INFERENCE_MODEL \
-  --env DEH_URL=$DEH_URL \
-  --env CHROMA_URL=$CHROMA_URL
+INFERENCE_MODEL=$INFERENCE_MODEL \
+DEH_URL=$DEH_URL \
+CHROMA_URL=$CHROMA_URL \
+llama stack run {{ name }} \
+  --port $LLAMA_STACK_PORT
 ```
 
 If you are using Llama Stack Safety / Shield APIs, use:
 
 ```bash
+INFERENCE_MODEL=$INFERENCE_MODEL \
+DEH_URL=$DEH_URL \
+SAFETY_MODEL=$SAFETY_MODEL \
+DEH_SAFETY_URL=$DEH_SAFETY_URL \
+CHROMA_URL=$CHROMA_URL \
 llama stack run ./run-with-safety.yaml \
-  --port $LLAMA_STACK_PORT \
-  --env INFERENCE_MODEL=$INFERENCE_MODEL \
-  --env DEH_URL=$DEH_URL \
-  --env SAFETY_MODEL=$SAFETY_MODEL \
-  --env DEH_SAFETY_URL=$DEH_SAFETY_URL \
-  --env CHROMA_URL=$CHROMA_URL
+  --port $LLAMA_STACK_PORT
 ```

llama_stack/distributions/meta-reference-gpu/doc_template.md

@@ -72,9 +72,9 @@ docker run \
   --gpu all \
   -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
   -v ~/.llama:/root/.llama \
+  -e INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct \
   llamastack/distribution-{{ name }} \
-  --port $LLAMA_STACK_PORT \
-  --env INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct
+  --port $LLAMA_STACK_PORT
 ```
 
 If you are using Llama Stack Safety / Shield APIs, use:
@@ -86,10 +86,10 @@ docker run \
   --gpu all \
   -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
   -v ~/.llama:/root/.llama \
+  -e INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct \
+  -e SAFETY_MODEL=meta-llama/Llama-Guard-3-1B \
   llamastack/distribution-{{ name }} \
-  --port $LLAMA_STACK_PORT \
-  --env INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct \
-  --env SAFETY_MODEL=meta-llama/Llama-Guard-3-1B
+  --port $LLAMA_STACK_PORT
 ```
 
 ### Via venv
@@ -98,16 +98,16 @@ Make sure you have done `uv pip install llama-stack` and have the Llama Stack CL
 
 ```bash
 llama stack build --distro {{ name }} --image-type venv
+INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct \
 llama stack run distributions/{{ name }}/run.yaml \
-  --port 8321 \
-  --env INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct
+  --port 8321
 ```
 
 If you are using Llama Stack Safety / Shield APIs, use:
 
 ```bash
+INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct \
+SAFETY_MODEL=meta-llama/Llama-Guard-3-1B \
 llama stack run distributions/{{ name }}/run-with-safety.yaml \
-  --port 8321 \
-  --env INFERENCE_MODEL=meta-llama/Llama-3.2-3B-Instruct \
-  --env SAFETY_MODEL=meta-llama/Llama-Guard-3-1B
+  --port 8321
 ```

llama_stack/distributions/nvidia/doc_template.md

@@ -118,10 +118,10 @@ docker run \
   --pull always \
   -p $LLAMA_STACK_PORT:$LLAMA_STACK_PORT \
   -v ./run.yaml:/root/my-run.yaml \
+  -e NVIDIA_API_KEY=$NVIDIA_API_KEY \
   llamastack/distribution-{{ name }} \
   --config /root/my-run.yaml \
-  --port $LLAMA_STACK_PORT \
-  --env NVIDIA_API_KEY=$NVIDIA_API_KEY
+  --port $LLAMA_STACK_PORT
 ```
 
 ### Via venv
@@ -131,10 +131,10 @@ If you've set up your local development environment, you can also build the imag
 ```bash
 INFERENCE_MODEL=meta-llama/Llama-3.1-8B-Instruct
 llama stack build --distro nvidia --image-type venv
+NVIDIA_API_KEY=$NVIDIA_API_KEY \
+INFERENCE_MODEL=$INFERENCE_MODEL \
 llama stack run ./run.yaml \
-  --port 8321 \
-  --env NVIDIA_API_KEY=$NVIDIA_API_KEY \
-  --env INFERENCE_MODEL=$INFERENCE_MODEL
+  --port 8321
 ```
 
 ## Example Notebooks