LCORE-2547: rename llama_stack modules and paths to ogx

AGENTS.md

@@ -203,7 +203,7 @@ src/
 │   │   └── redaction/                            # PII redaction capability for Pydantic AI agents
 │   │       ├── capability.py                     # Pydantic AI capability for PII redaction of model messages
 │   │       └── core.py                           # Core redaction logic for PII detection and replacement
-│   └── llamastack/                               # Pydantic AI provider for OGX
+│   └── ogx/                                      # Pydantic AI provider for OGX
 │       ├── _model.py                             # Custom OpenAI Responses model that works around OGX streaming quirks
 │       ├── _provider.py                          # OGX provider implementation for Pydantic AI
 │       └── _transport.py                         # httpx transport that routes OpenAI-compatible requests through an OGX library client
@@ -224,7 +224,7 @@ src/
 │   ├── degraded_mode.py                          # Degraded mode state tracking
 │   ├── endpoints.py                              # Utility functions for endpoint handlers
 │   ├── json_schema_updater.py                    # Function to transform a JSON Schema-like dictionary into an OpenAPI-compatible schema
-│   ├── llama_stack_version.py                    # Check if the OGX version is supported by the LCS
+│   ├── ogx_version.py                            # Check if the OGX version is supported by the LCS
 │   ├── markdown_repair.py                        # Utilities for repairing truncated markdown content
 │   ├── mcp_auth_headers.py                       # Utilities for resolving MCP server authorization headers
 │   ├── mcp_headers.py                            # MCP headers handling
@@ -250,7 +250,7 @@ src/
 │   └── vector_search.py                          # Vector search utilities for query endpoints
 ├── sentry.py                                     # Sentry error tracking initialization and configuration
 ├── lightspeed_stack.py                           # Entry point to the Lightspeed Core Stack REST API service
-├── llama_stack_configuration.py                  # OGX configuration enrichment and synthesis
+├── ogx_configuration.py                          # OGX configuration enrichment and synthesis
 ├── log.py                                        # Log utilities
 ├── client.py                                     # OGX client wrapper (Singleton)
 ├── configuration.py                              # Config management (Singleton)

Makefile

@@ -48,7 +48,7 @@ build-ogx-image: remove-ogx-container ## Build OGX container image
 		echo "ERROR: No container runtime found. Install podman or docker."; \
 		exit 1; \
 	fi
-	$(CONTAINER_RUNTIME) build -f deploy/llama-stack/test.containerfile -t $(OGX_IMAGE) .
+	$(CONTAINER_RUNTIME) build -f deploy/ogx/test.containerfile -t $(OGX_IMAGE) .
 
 stop-ogx-container: ## Gracefully stop OGX container
 	@if [ -n "$(CONTAINER_RUNTIME)" ] && $(CONTAINER_RUNTIME) inspect $(OGX_CONTAINER_NAME) >/dev/null 2>&1; then \
@@ -84,8 +84,8 @@ start-ogx-container: build-ogx-image ## Start OGX container
 		--health-start-period 20s \
 		-v $(PWD)/$(OGX_CONFIG):/opt/app-root/run.yaml:z \
 		-v $(PWD)/$(CONFIG):/opt/app-root/lightspeed-stack.yaml:ro,z \
-		-v $(PWD)/scripts/llama-stack-entrypoint.sh:/opt/app-root/enrich-entrypoint.sh:ro,z \
-		-v $(PWD)/src/llama_stack_configuration.py:/opt/app-root/llama_stack_configuration.py:ro,z \
+		-v $(PWD)/scripts/ogx-entrypoint.sh:/opt/app-root/enrich-entrypoint.sh:ro,z \
+		-v $(PWD)/src/ogx_configuration.py:/opt/app-root/ogx_configuration.py:ro,z \
 		-e OPENAI_API_KEY \
 		-e BRAVE_SEARCH_API_KEY \
 		-e TAVILY_SEARCH_API_KEY \
@@ -143,7 +143,7 @@ clean-ogx: remove-ogx-container ## Remove container and image
 	fi
 
 run-ogx-local: ## Start OGX with enriched config (for local service mode)
-	uv run src/llama_stack_configuration.py -c $(CONFIG) -i $(OGX_CONFIG) -o $(OGX_CONFIG) && \
+	uv run src/ogx_configuration.py -c $(CONFIG) -i $(OGX_CONFIG) -o $(OGX_CONFIG) && \
 	uv run ogx stack run $(OGX_CONFIG)
 
 test-unit: ## Run the unit tests

deploy/llama-stack/README.md → deploy/ogx/README.md

@@ -5,13 +5,13 @@
 Llama Stack distribution itself, the image bundles the pieces needed to
 generate its run configuration at container start:
 
-- `/opt/app-root/llama_stack_configuration.py` — the config-generation
-  script (copied from `src/llama_stack_configuration.py`).
+- `/opt/app-root/ogx_configuration.py` — the config-generation
+  script (copied from `src/ogx_configuration.py`).
 - `/opt/app-root/data/default_run.yaml` — the shipped default baseline
   (copied from `src/data/`), resolved by the script as `./data/` relative
   to its own location.
 - `/opt/app-root/enrich-entrypoint.sh` — the entrypoint (copied from
-  `scripts/llama-stack-entrypoint.sh`).
+  `scripts/ogx-entrypoint.sh`).
 
 ## Startup modes
 
@@ -36,7 +36,7 @@ services:
   llama-stack:
     build:
       context: .
-      dockerfile: deploy/llama-stack/test.containerfile
+      dockerfile: deploy/ogx/test.containerfile
     ports:
       - "8321:8321"
     volumes:

deploy/llama-stack/test.containerfile → deploy/ogx/test.containerfile

@@ -42,12 +42,12 @@ RUN mkdir -p /opt/app-root/src/.llama/storage \
 # Copy the config-generation script for runtime synthesis (unified mode) or
 # enrichment (legacy mode), plus the shipped default baseline it resolves as
 # ./data/default_run.yaml relative to its own location (load_default_baseline)
-COPY src/llama_stack_configuration.py /opt/app-root/llama_stack_configuration.py
+COPY src/ogx_configuration.py /opt/app-root/ogx_configuration.py
 COPY src/data /opt/app-root/data
-COPY scripts/llama-stack-entrypoint.sh /opt/app-root/enrich-entrypoint.sh
+COPY scripts/ogx-entrypoint.sh /opt/app-root/enrich-entrypoint.sh
 RUN chmod +x /opt/app-root/enrich-entrypoint.sh && \
     chown -R 1001:0 /opt/app-root/enrich-entrypoint.sh \
-        /opt/app-root/llama_stack_configuration.py /opt/app-root/data
+        /opt/app-root/ogx_configuration.py /opt/app-root/data
 
 # Switch back to the original user
 USER 1001

docker-compose.yaml

@@ -3,7 +3,7 @@ services:
   llama-stack:
     build:
       context: .
-      dockerfile: deploy/llama-stack/test.containerfile
+      dockerfile: deploy/ogx/test.containerfile
     platform: linux/amd64
     container_name: llama-stack
     ports:
@@ -18,8 +18,8 @@ services:
       # from lightspeed-stack.yaml and this mount is ignored
       - ./run.yaml:/opt/app-root/run.yaml:z
       # Host copies so `docker compose up` picks up script changes without rebuilding OGX
-      - ./scripts/llama-stack-entrypoint.sh:/opt/app-root/enrich-entrypoint.sh:ro,z
-      - ./src/llama_stack_configuration.py:/opt/app-root/llama_stack_configuration.py:ro,z
+      - ./scripts/ogx-entrypoint.sh:/opt/app-root/enrich-entrypoint.sh:ro,z
+      - ./src/ogx_configuration.py:/opt/app-root/ogx_configuration.py:ro,z
       - ./src/data:/opt/app-root/data:ro,z
       - ${GCP_KEYS_PATH:-./tmp/.gcp-keys-dummy}:/opt/app-root/.gcp-keys:ro
       - ./lightspeed-stack.yaml:/opt/app-root/lightspeed-stack.yaml:ro,z

docs/devel_doc/container_orchestration.md

@@ -83,7 +83,7 @@ When you run `make run`, the following happens:
 make build-llama-stack-image
 ```
 
-- Builds from `deploy/llama-stack/test.containerfile`
+- Builds from `deploy/ogx/test.containerfile`
 - Tags as `lightspeed-llama-stack:local` (customizable via `LLAMA_STACK_IMAGE`)
 - Only rebuilds if the image doesn't exist or source files changed
 - Removes any existing container before building (ensures clean build)
@@ -459,7 +459,7 @@ Container logs:
 3. **Test config enrichment:**
    ```bash
    # Run enrichment script manually to check for errors
-   uv run src/llama_stack_configuration.py \
+   uv run src/ogx_configuration.py \
      -c lightspeed-stack.yaml \
      -i run.yaml \
      -o /tmp/enriched-run.yaml
@@ -649,7 +649,7 @@ This grants read-only access to UID 1001 (container user) without changing base
 **macOS note:** macOS uses BSD ACLs and cannot assign numeric UID-based ACLs to non-existent host users. If you are testing locally on macOS, you must temporarily use `chmod 644` to allow the container access, but **be aware that this makes the credentials file world-readable on your host machine.** Alternately, ensure your local user matches the container's execution environment.
 
 **Why this happens:**
-This is expected container behavior. The container runs as a non-root user (UID 1001) for security - see `USER 1001` in `deploy/llama-stack/test.containerfile`. Files with `600` permissions are only accessible to their owner, and the container's UID differs from your host UID.
+This is expected container behavior. The container runs as a non-root user (UID 1001) for security - see `USER 1001` in `deploy/ogx/test.containerfile`. Files with `600` permissions are only accessible to their owner, and the container's UID differs from your host UID.
 
 **Production recommendation:**
 For production deployments, avoid mounting credential files entirely. Instead use:
@@ -683,9 +683,9 @@ When the OGX container starts, it automatically enriches the `run.yaml` file wit
 
 #### How It Works
 
-1. **Entrypoint script** (`scripts/llama-stack-entrypoint.sh`) is mounted at `/opt/app-root/enrich-entrypoint.sh`
-2. **Script runs** `/opt/app-root/.venv/bin/python3 /opt/app-root/llama_stack_configuration.py`
-3. **Enrichment logic** (`src/llama_stack_configuration.py`) reads both configs and merges them
+1. **Entrypoint script** (`scripts/ogx-entrypoint.sh`) is mounted at `/opt/app-root/enrich-entrypoint.sh`
+2. **Script runs** `/opt/app-root/.venv/bin/python3 /opt/app-root/ogx_configuration.py`
+3. **Enrichment logic** (`src/ogx_configuration.py`) reads both configs and merges them
 4. **Output** is written to `/tmp/enriched-run.yaml` inside the container
 5. **OGX starts** with the enriched config
 
@@ -699,7 +699,7 @@ When the OGX container starts, it automatically enriches the `run.yaml` file wit
 
 ```bash
 # Run enrichment locally to see output
-uv run src/llama_stack_configuration.py \
+uv run src/ogx_configuration.py \
   -c lightspeed-stack.yaml \
   -i run.yaml \
   -o enriched-run.yaml
@@ -716,8 +716,8 @@ The container uses these volume mounts:
 |-----------|----------------|------|---------|
 | `$(PWD)/run.yaml` | `/opt/app-root/run.yaml` | rw | OGX config (enriched version written here) |
 | `$(PWD)/lightspeed-stack.yaml` | `/opt/app-root/lightspeed-stack.yaml` | ro | LCORE config (read for enrichment) |
-| `$(PWD)/scripts/llama-stack-entrypoint.sh` | `/opt/app-root/enrich-entrypoint.sh` | ro | Entrypoint script with enrichment logic |
-| `$(PWD)/src/llama_stack_configuration.py` | `/opt/app-root/llama_stack_configuration.py` | ro | Python enrichment script |
+| `$(PWD)/scripts/ogx-entrypoint.sh` | `/opt/app-root/enrich-entrypoint.sh` | ro | Entrypoint script with enrichment logic |
+| `$(PWD)/src/ogx_configuration.py` | `/opt/app-root/ogx_configuration.py` | ro | Python enrichment script |
 
 **SELinux labels:**
 - `:z`: Relabels for sharing between host and container (read-write)
@@ -734,7 +734,7 @@ If you need more control than the Makefile provides, you can manage the containe
 
 #### Build the Image
 ```bash
-podman build -f deploy/llama-stack/test.containerfile -t my-llama-stack:custom .
+podman build -f deploy/ogx/test.containerfile -t my-llama-stack:custom .
 ```
 
 #### Run the Container

docs/user_doc/deployment_guide.md

@@ -170,7 +170,7 @@ and `library_client_config_path` must not be set (unified and legacy inputs
 are mutually exclusive).
 
 The reference profiles are sanity-checked by the unit suite
-(`tests/unit/test_llama_stack_synthesize.py`), so they stay loadable as the
+(`tests/unit/test_ogx_synthesize.py`), so they stay loadable as the
 synthesizer evolves.
 
 

scripts/llama-stack-entrypoint.sh → scripts/ogx-entrypoint.sh

@@ -2,7 +2,7 @@
 # Entrypoint for OGX container.
 #
 # Generates the run configuration from the mounted lightspeed-stack.yaml and
-# starts OGX. The Python CLI (llama_stack_configuration.py) auto-detects the
+# starts OGX. The Python CLI (ogx_configuration.py) auto-detects the
 # configuration shape:
 #   - unified mode: the lightspeed config carries a synthesis input (a
 #     non-empty inference.providers or vector_store.providers, or a
@@ -23,7 +23,7 @@ LIGHTSPEED_CONFIG="${LIGHTSPEED_CONFIG:-/opt/app-root/lightspeed-stack.yaml}"
 if [ -f "$LIGHTSPEED_CONFIG" ]; then
     echo "Generating llama-stack config from $LIGHTSPEED_CONFIG (mode auto-detected)..."
     GENERATION_FAILED=0
-    /opt/app-root/.venv/bin/python3 /opt/app-root/llama_stack_configuration.py \
+    /opt/app-root/.venv/bin/python3 /opt/app-root/ogx_configuration.py \
         -c "$LIGHTSPEED_CONFIG" \
         -i "$INPUT_CONFIG" \
         -o "$GENERATED_CONFIG" 2>&1 || GENERATION_FAILED=1

scripts/llama_stack_tutorial.sh → scripts/ogx_tutorial.sh

@@ -319,9 +319,9 @@ Next Steps:
 4. Experiment with different models and tools
 
 Resources:
-- This tutorial script: ./llama_stack_tutorial.sh
-- Run without pauses: ./llama_stack_tutorial.sh --no-wait
-- Python version: ./llama_stack_tutorial.py (requires: uv run python3)
+- This tutorial script: ./ogx_tutorial.sh
+- Run without pauses: ./ogx_tutorial.sh --no-wait
+- Python version: ./ogx_tutorial.py (requires: uv run python3)
 - Interactive docs: http://localhost:8321/docs
 
 Happy exploring! 🚀

src/README.md

@@ -16,7 +16,7 @@ Constants used in business logic.
 
 Entry point to the Lightspeed Core Stack REST API service.
 
-## [llama_stack_configuration.py](llama_stack_configuration.py)
+## [ogx_configuration.py](ogx_configuration.py)
 
 OGX configuration enrichment and synthesis.
 

src/app/main.py

@@ -27,7 +27,7 @@
 from models.api.responses.error import InternalServerErrorResponse
 from observability.sentry import initialize_sentry
 from utils.degraded_mode import DegradedModeTracker
-from utils.llama_stack_version import check_ogx_version
+from utils.ogx_version import check_ogx_version
 
 logger = get_logger(__name__)
 

src/client/ogx.py

@@ -15,16 +15,16 @@
 import constants
 from authorization.azure_token_manager import AzureEntraIDManager
 from configuration import configuration
-from llama_stack_configuration import (
+from log import get_logger, setup_logging
+from models.api.responses.error import ServiceUnavailableResponse
+from models.config import OgxConfiguration
+from ogx_configuration import (
     YamlDumper,
     enrich_azure_entra_id_inference,
     enrich_byok_rag,
     enrich_solr,
     synthesize_to_file,
 )
-from log import get_logger, setup_logging
-from models.api.responses.error import ServiceUnavailableResponse
-from models.config import OgxConfiguration
 from utils.model_list import parse_model_list_response
 from utils.types import Singleton
 

src/constants.py

@@ -202,7 +202,7 @@
 CACHE_TYPE_NOOP: Final[str] = "noop"
 
 # BYOK RAG
-# Backends that have enrichment support in llama_stack_configuration.py
+# Backends that have enrichment support in ogx_configuration.py
 SUPPORTED_RAG_BACKENDS: Final[frozenset[str]] = frozenset({"faiss", "pgvector"})
 
 # Default RAG backend for bring-your-own-knowledge RAG configurations

src/data/__init__.py

@@ -2,7 +2,7 @@
 
 Currently holds ``default_run.yaml``, the built-in baseline OGX
 configuration used by unified-mode synthesis (see
-``llama_stack_configuration.load_default_baseline``). Making this directory a
+``ogx_configuration.load_default_baseline``). Making this directory a
 package ensures the data file is included in built wheels and resolvable both
 in editable installs and from site-packages.
 """

src/data/default_run.yaml

@@ -2,7 +2,7 @@
 #
 # This file is the starting point when a unified `lightspeed-stack.yaml`
 # selects `llama_stack.config.baseline: default` (the default) and does not
-# point at a `profile:`. The synthesizer (src/llama_stack_configuration.py)
+# point at a `profile:`. The synthesizer (src/ogx_configuration.py)
 # layers enrichment, high-level `inference.providers`, and `native_override`
 # on top of this baseline to produce the final run.yaml handed to OGX.
 #

src/lightspeed_stack.py

@@ -10,8 +10,8 @@
 import constants
 from configuration import configuration
 from constants import LIGHTSPEED_STACK_LOG_LEVEL_ENV_VAR
-from llama_stack_configuration import migrate_config_dumb
 from log import get_logger, setup_logging
+from ogx_configuration import migrate_config_dumb
 from runners.quota_scheduler import start_quota_scheduler
 from runners.uvicorn import start_uvicorn
 from utils.dumpers import config_dumper, models_dumper

src/llama_stack_configuration.py → src/ogx_configuration.py

@@ -1,8 +1,8 @@
 """OGX configuration enrichment and synthesis.
 
 This module can be used in two ways:
-1. As a script: `python llama_stack_configuration.py -c config.yaml`
-2. As a module: `from llama_stack_configuration import generate_configuration`
+1. As a script: `python ogx_configuration.py -c config.yaml`
+2. As a module: `from ogx_configuration import generate_configuration`
 
 Two related responsibilities live here:
 

src/pydantic_ai_lightspeed/capabilities/question_validity/_capability.py

@@ -40,7 +40,7 @@
     QuestionValidityConfig,
 )
 from pydantic_ai_lightspeed.capabilities.base import AbstractSafetyCapability
-from pydantic_ai_lightspeed.llamastack import OgxResponsesModel
+from pydantic_ai_lightspeed.ogx import OgxResponsesModel
 from utils.conversations import append_turn_to_conversation
 
 logger = get_logger(__name__)

src/pydantic_ai_lightspeed/llamastack/README.md → src/pydantic_ai_lightspeed/ogx/README.md

@@ -1,4 +1,4 @@
-# List of source files stored in `src/pydantic_ai_lightspeed/llamastack` directory
+# List of source files stored in `src/pydantic_ai_lightspeed/ogx` directory
 
 ## [__init__.py](__init__.py)
 

src/pydantic_ai_lightspeed/ogx/__init__.py

@@ -0,0 +1,6 @@
+"""Pydantic AI provider for OGX."""
+
+from pydantic_ai_lightspeed.ogx._model import OgxResponsesModel
+from pydantic_ai_lightspeed.ogx._provider import OgxProvider
+
+__all__ = ["OgxProvider", "OgxResponsesModel"]

src/pydantic_ai_lightspeed/llamastack/_model.py → src/pydantic_ai_lightspeed/ogx/_model.py

@@ -48,7 +48,7 @@
 
 from log import get_logger
 from models.common.responses.responses_api_params import ResponsesApiParams
-from pydantic_ai_lightspeed.llamastack._provider import OgxProvider
+from pydantic_ai_lightspeed.ogx._provider import OgxProvider
 
 logger = get_logger(__name__)
 

src/pydantic_ai_lightspeed/llamastack/_provider.py → src/pydantic_ai_lightspeed/ogx/_provider.py

@@ -14,7 +14,7 @@
 from pydantic_ai.profiles.openai import openai_model_profile
 from pydantic_ai.providers import Provider
 
-from pydantic_ai_lightspeed.llamastack._transport import (
+from pydantic_ai_lightspeed.ogx._transport import (
     OgxLibraryTransport,
     wrap_http_client_with_provider_data,
 )

src/utils/README.md

@@ -48,7 +48,7 @@ Input sanitization to detect and block obfuscated prompt injection attempts.
 
 Function to transform a JSON Schema-like dictionary into an OpenAPI-compatible schema.
 
-## [llama_stack_version.py](llama_stack_version.py)
+## [ogx_version.py](ogx_version.py)
 
 Check if the OGX version is supported by the LCS.
 

src/utils/pydantic_ai_helpers.py

@@ -23,7 +23,7 @@
 )
 from pydantic_ai_lightspeed.capabilities import QuestionValidity
 from pydantic_ai_lightspeed.capabilities.redaction import PiiRedactionCapability
-from pydantic_ai_lightspeed.llamastack import OgxResponsesModel
+from pydantic_ai_lightspeed.ogx import OgxResponsesModel
 from utils.shields import get_shields_for_request
 
 _AGENT_SKILLS_PROVIDER_ID: Final[str] = "agent-skills"