lightspeed-core · tisnik · Aug 3, 2026 · Jul 22, 2026 · Jul 22, 2026 · Jul 22, 2026
diff --git a/.konflux/requirements.hashes.source.txt b/.konflux/requirements.hashes.source.txt
@@ -99,15 +99,6 @@ google-cloud-bigquery==3.42.2 \
 google-cloud-resource-manager==1.18.0 \
     --hash=sha256:69bab144acd75878ebe44b720903dc3d140cb7d3be3261eaa8bc81e48afaff33 \
     --hash=sha256:db689f800a14c66d041196a7fbb8bb8aae8dc87f28c2929e101a5ec766b15512
-llama-stack==0.6.0 \
-    --hash=sha256:b804830664dc91e54c7225a7a081cb1874c48fc18573569c19fac4a9397e8076 \
-    --hash=sha256:d92711791633f5505a4473ffba3f3e26acb700716fddab5aec419d99e614c802
-llama-stack-api==0.6.0 \
-    --hash=sha256:b99a03aba3659736b6b540c9e5e674b1daac2bf5eeb2a68795113d62b8250672 \
-    --hash=sha256:f0f3a1a6239a5d3b8c7ef02cefdf817c96c6461dcd8a82c1689ac67ec3107270
-llama-stack-client==0.6.0 \
-    --hash=sha256:3290aac36dcafbd1bc0baaf995522e2037f57056672b5a1516af112a4210f3ea \
-    --hash=sha256:7e514a6ffd92f237aceb062dadc4db44e24a3cd9c4ea35e25173d1e0739beb8e
 oci==2.182.1 \
     --hash=sha256:0c616a6bc3bc458464bc3456469d8da63a1a2d2277e9314b41a1c4e76d5df523 \
     --hash=sha256:9862de221f2abe9cf8319393eec58ea59c014fd9b61afaf0a3cca163e2a508b0

diff --git a/.konflux/requirements.hashes.wheel.txt b/.konflux/requirements.hashes.wheel.txt
@@ -222,6 +222,12 @@ oauthlib==3.3.1 \
     --hash=sha256:c6fbab4f1a77a539f01175e2ea74b9552806bd0a849c70e744fda4ab801031c0
 openai==2.44.0 \
     --hash=sha256:87429e9a4d15b2918a03b040b6330fa03fc175bbf0bc6eaff7ae61c93cd42c53
+ogx==1.0.2+rhaiv.0 \
+    --hash=sha256:52c891af9dfb22f884d2dae150e5e94d04e492f452ec811bbc61ba84257de000
+ogx-api==1.0.2+rhaiv.0 \
+    --hash=sha256:265dff54f2367f4f366952e7a17abe06ceeef777a809259a5fba4715f4df4e90
+ogx-client==1.0.2 \
+    --hash=sha256:3ac249fb8365a4cb801815c95113fc6954a16193db5247d9b9f053398a80add9
 opentelemetry-api==1.42.1 \
     --hash=sha256:078a23234520ddf8654a48045b8f582c4a73c9bd5da2f0d23e05aba6e40a7c91
 opentelemetry-distro==0.63b1 \
@@ -374,6 +380,8 @@ sse-starlette==3.4.5 \
     --hash=sha256:c1f701f19f43c181be2d62c6164e90b7e8e4d5d67956fface4c794ac7a784962
 starlette==1.3.1 \
     --hash=sha256:19acd5bfb037734bb027cb753f36379bca5f72da9b79fa7fb3ef4639b1999f6b
+structlog==26.1.0 \
+    --hash=sha256:0054592356113edd68e13e9be1be5062566ddbf15a24bb3f400ab764a26b0b36
 sympy==1.14.0 \
     --hash=sha256:6cf6a1e379dbba75ca560e2cbd4394144f9e7e7335ee1be872c224f004eeb172
 tenacity==9.1.4 \
@@ -439,3 +447,6 @@ yarl==1.24.2 \
     --hash=sha256:c72aabbf544bceb3fa895376a188a90b7152efd47045b589fd358505ad8a85d4
 zipp==4.1.0 \
     --hash=sha256:f04782919048b93bac92fa1d20b2dd8a062b30ebee718e834cf663874884c645
+zstandard==0.25.0 \
+    --hash=sha256:3f12f3143d2dba84081a79932a30e8179df5c655134d1ae613cc760aab9e4ff2 \
+    --hash=sha256:c734ffe543d7aaf7ff9848fb06fcc4c251e2c1670eab845b8983b6ff24f993fd
diff --git a/.tekton/lightspeed-stack-0-7-pull-request.yaml b/.tekton/lightspeed-stack-0-7-pull-request.yaml
@@ -53,7 +53,7 @@ spec:
            ],
            "requirements_build_files": ["requirements-build.txt"],
            "binary": {
-             "packages": "a2a-sdk,accelerate,aiofile,aiohappyeyeballs,aiohttp,aiosignal,aiosqlite,annotated-doc,annotated-types,anthropic,anyio,argcomplete,asyncpg,attrs,authlib,autoevals,azure-core,azure-identity,beartype,cachetools,caio,certifi,cffi,chardet,charset-normalizer,chevron,click,cryptography,datasets,dill,distro,dnspython,docstring-parser,durationpy,einops,email-validator,emoji,exceptiongroup,executing,faiss-cpu,fastapi,fastmcp-slim,fastuuid,filelock,fire,frozenlist,fsspec,genai-prices,google-api-core,google-auth,google-cloud-core,google-cloud-storage,google-crc32c,google-genai,google-resumable-media,googleapis-common-protos,greenlet,griffelib,grpc-google-iam-v1,grpcio,grpcio-status,h11,hf-xet,httpcore,httpcore2,httpx,httpx-sse,httpx2,huggingface-hub,idna,importlib-metadata,jaraco-classes,jaraco-context,jaraco-functools,jeepney,jinja2,jiter,joblib,joserfc,jsonpath-ng,jsonschema,jsonschema-specifications,keyring,kubernetes,langdetect,litellm,logfire,logfire-api,markdown-it-py,markupsafe,maturin,mcp,mdurl,more-itertools,mpmath,msal,msal-extensions,multidict,multiprocess,narwhals,networkx,nltk,numpy,oauthlib,openai,opentelemetry-api,opentelemetry-distro,opentelemetry-exporter-otlp,opentelemetry-exporter-otlp-proto-common,opentelemetry-exporter-otlp-proto-grpc,opentelemetry-exporter-otlp-proto-http,opentelemetry-instrumentation,opentelemetry-instrumentation-httpx,opentelemetry-proto,opentelemetry-sdk,opentelemetry-semantic-conventions,opentelemetry-util-http,oracledb,packaging,pandas,peft,pip,platformdirs,polyleven,prometheus-client,prompt-toolkit,propcache,proto-plus,protobuf,psutil,psycopg2-binary,py-key-value-aio,pyaml,pyarrow,pyasn1,pyasn1-modules,pycparser,pydantic,pydantic-ai,pydantic-ai-slim,pydantic-core,pydantic-evals,pydantic-graph,pydantic-settings,pygments,pyjwt,pyopenssl,pypdf,pyperclip,python-dateutil,python-dotenv,python-multipart,pytz,pyyaml,referencing,regex,requests,requests-oauthlib,rich,rpds-py,safetensors,scikit-learn,scipy,secretstorage,semver,sentence-transformers,sentry-sdk,setuptools,shellingham,six,sniffio,sqlalchemy,sse-starlette,starlette,sympy,tenacity,termcolor,threadpoolctl,tiktoken,tokenizers,torch,tornado,tqdm,transformers,tree-sitter,triton,trl,truststore,typer,typing-extensions,typing-inspection,urllib3,uv,uv-build,uvicorn,wcwidth,websocket-client,websockets,wrapt,xxhash,yarl,zipp",
+             "packages": "a2a-sdk,accelerate,aiofile,aiohappyeyeballs,aiohttp,aiosignal,aiosqlite,annotated-doc,annotated-types,anthropic,anyio,argcomplete,asyncpg,attrs,authlib,autoevals,azure-core,azure-identity,beartype,cachetools,caio,certifi,cffi,chardet,charset-normalizer,chevron,click,cryptography,datasets,dill,distro,dnspython,docstring-parser,durationpy,einops,email-validator,emoji,exceptiongroup,executing,faiss-cpu,fastapi,fastmcp-slim,fastuuid,filelock,fire,frozenlist,fsspec,genai-prices,google-api-core,google-auth,google-cloud-core,google-cloud-storage,google-crc32c,google-genai,google-resumable-media,googleapis-common-protos,greenlet,griffelib,grpc-google-iam-v1,grpcio,grpcio-status,h11,hf-xet,httpcore,httpcore2,httpx,httpx-sse,httpx2,huggingface-hub,idna,importlib-metadata,jaraco-classes,jaraco-context,jaraco-functools,jeepney,jinja2,jiter,joblib,joserfc,jsonpath-ng,jsonschema,jsonschema-specifications,keyring,kubernetes,langdetect,litellm,logfire,logfire-api,markdown-it-py,markupsafe,maturin,mcp,mdurl,more-itertools,mpmath,msal,msal-extensions,multidict,multiprocess,narwhals,networkx,nltk,numpy,oauthlib,ogx,ogx-api,ogx-client,openai,opentelemetry-api,opentelemetry-distro,opentelemetry-exporter-otlp,opentelemetry-exporter-otlp-proto-common,opentelemetry-exporter-otlp-proto-grpc,opentelemetry-exporter-otlp-proto-http,opentelemetry-instrumentation,opentelemetry-instrumentation-httpx,opentelemetry-proto,opentelemetry-sdk,opentelemetry-semantic-conventions,opentelemetry-util-http,oracledb,packaging,pandas,peft,pip,platformdirs,polyleven,prometheus-client,prompt-toolkit,propcache,proto-plus,protobuf,psutil,psycopg2-binary,py-key-value-aio,pyaml,pyarrow,pyasn1,pyasn1-modules,pycparser,pydantic,pydantic-ai,pydantic-ai-slim,pydantic-core,pydantic-evals,pydantic-graph,pydantic-settings,pygments,pyjwt,pyopenssl,pypdf,pyperclip,python-dateutil,python-dotenv,python-multipart,pytz,pyyaml,referencing,regex,requests,requests-oauthlib,rich,rpds-py,safetensors,scikit-learn,scipy,secretstorage,semver,sentence-transformers,sentry-sdk,setuptools,shellingham,six,sniffio,sqlalchemy,sse-starlette,starlette,structlog,sympy,tenacity,termcolor,threadpoolctl,tiktoken,tokenizers,torch,tornado,tqdm,transformers,tree-sitter,triton,trl,truststore,typer,typing-extensions,typing-inspection,urllib3,uv,uv-build,uvicorn,wcwidth,websocket-client,websockets,wrapt,xxhash,yarl,zipp,zstandard",
              "os": "linux",
              "arch": "x86_64,aarch64",
              "py_version": 312

diff --git a/.tekton/lightspeed-stack-0-7-push.yaml b/.tekton/lightspeed-stack-0-7-push.yaml
@@ -54,7 +54,7 @@ spec:
           ],
           "requirements_build_files": ["requirements-build.txt"],
           "binary": {
-            "packages": "a2a-sdk,accelerate,aiofile,aiohappyeyeballs,aiohttp,aiosignal,aiosqlite,annotated-doc,annotated-types,anthropic,anyio,argcomplete,asyncpg,attrs,authlib,autoevals,azure-core,azure-identity,beartype,cachetools,caio,certifi,cffi,chardet,charset-normalizer,chevron,click,cryptography,datasets,dill,distro,dnspython,docstring-parser,durationpy,einops,email-validator,emoji,exceptiongroup,executing,faiss-cpu,fastapi,fastmcp-slim,fastuuid,filelock,fire,frozenlist,fsspec,genai-prices,google-api-core,google-auth,google-cloud-core,google-cloud-storage,google-crc32c,google-genai,google-resumable-media,googleapis-common-protos,greenlet,griffelib,grpc-google-iam-v1,grpcio,grpcio-status,h11,hf-xet,httpcore,httpcore2,httpx,httpx-sse,httpx2,huggingface-hub,idna,importlib-metadata,jaraco-classes,jaraco-context,jaraco-functools,jeepney,jinja2,jiter,joblib,joserfc,jsonpath-ng,jsonschema,jsonschema-specifications,keyring,kubernetes,langdetect,litellm,logfire,logfire-api,markdown-it-py,markupsafe,maturin,mcp,mdurl,more-itertools,mpmath,msal,msal-extensions,multidict,multiprocess,narwhals,networkx,nltk,numpy,oauthlib,openai,opentelemetry-api,opentelemetry-distro,opentelemetry-exporter-otlp,opentelemetry-exporter-otlp-proto-common,opentelemetry-exporter-otlp-proto-grpc,opentelemetry-exporter-otlp-proto-http,opentelemetry-instrumentation,opentelemetry-instrumentation-httpx,opentelemetry-proto,opentelemetry-sdk,opentelemetry-semantic-conventions,opentelemetry-util-http,oracledb,packaging,pandas,peft,pip,platformdirs,polyleven,prometheus-client,prompt-toolkit,propcache,proto-plus,protobuf,psutil,psycopg2-binary,py-key-value-aio,pyaml,pyarrow,pyasn1,pyasn1-modules,pycparser,pydantic,pydantic-ai,pydantic-ai-slim,pydantic-core,pydantic-evals,pydantic-graph,pydantic-settings,pygments,pyjwt,pyopenssl,pypdf,pyperclip,python-dateutil,python-dotenv,python-multipart,pytz,pyyaml,referencing,regex,requests,requests-oauthlib,rich,rpds-py,safetensors,scikit-learn,scipy,secretstorage,semver,sentence-transformers,sentry-sdk,setuptools,shellingham,six,sniffio,sqlalchemy,sse-starlette,starlette,sympy,tenacity,termcolor,threadpoolctl,tiktoken,tokenizers,torch,tornado,tqdm,transformers,tree-sitter,triton,trl,truststore,typer,typing-extensions,typing-inspection,urllib3,uv,uv-build,uvicorn,wcwidth,websocket-client,websockets,wrapt,xxhash,yarl,zipp",
+            "packages": "a2a-sdk,accelerate,aiofile,aiohappyeyeballs,aiohttp,aiosignal,aiosqlite,annotated-doc,annotated-types,anthropic,anyio,argcomplete,asyncpg,attrs,authlib,autoevals,azure-core,azure-identity,beartype,cachetools,caio,certifi,cffi,chardet,charset-normalizer,chevron,click,cryptography,datasets,dill,distro,dnspython,docstring-parser,durationpy,einops,email-validator,emoji,exceptiongroup,executing,faiss-cpu,fastapi,fastmcp-slim,fastuuid,filelock,fire,frozenlist,fsspec,genai-prices,google-api-core,google-auth,google-cloud-core,google-cloud-storage,google-crc32c,google-genai,google-resumable-media,googleapis-common-protos,greenlet,griffelib,grpc-google-iam-v1,grpcio,grpcio-status,h11,hf-xet,httpcore,httpcore2,httpx,httpx-sse,httpx2,huggingface-hub,idna,importlib-metadata,jaraco-classes,jaraco-context,jaraco-functools,jeepney,jinja2,jiter,joblib,joserfc,jsonpath-ng,jsonschema,jsonschema-specifications,keyring,kubernetes,langdetect,litellm,logfire,logfire-api,markdown-it-py,markupsafe,maturin,mcp,mdurl,more-itertools,mpmath,msal,msal-extensions,multidict,multiprocess,narwhals,networkx,nltk,numpy,oauthlib,ogx,ogx-api,ogx-client,openai,opentelemetry-api,opentelemetry-distro,opentelemetry-exporter-otlp,opentelemetry-exporter-otlp-proto-common,opentelemetry-exporter-otlp-proto-grpc,opentelemetry-exporter-otlp-proto-http,opentelemetry-instrumentation,opentelemetry-instrumentation-httpx,opentelemetry-proto,opentelemetry-sdk,opentelemetry-semantic-conventions,opentelemetry-util-http,oracledb,packaging,pandas,peft,pip,platformdirs,polyleven,prometheus-client,prompt-toolkit,propcache,proto-plus,protobuf,psutil,psycopg2-binary,py-key-value-aio,pyaml,pyarrow,pyasn1,pyasn1-modules,pycparser,pydantic,pydantic-ai,pydantic-ai-slim,pydantic-core,pydantic-evals,pydantic-graph,pydantic-settings,pygments,pyjwt,pyopenssl,pypdf,pyperclip,python-dateutil,python-dotenv,python-multipart,pytz,pyyaml,referencing,regex,requests,requests-oauthlib,rich,rpds-py,safetensors,scikit-learn,scipy,secretstorage,semver,sentence-transformers,sentry-sdk,setuptools,shellingham,six,sniffio,sqlalchemy,sse-starlette,starlette,structlog,sympy,tenacity,termcolor,threadpoolctl,tiktoken,tokenizers,torch,tornado,tqdm,transformers,tree-sitter,triton,trl,truststore,typer,typing-extensions,typing-inspection,urllib3,uv,uv-build,uvicorn,wcwidth,websocket-client,websockets,wrapt,xxhash,yarl,zipp,zstandard",
             "os": "linux",
             "arch": "x86_64,aarch64",
             "py_version": 312

diff --git a/AGENTS.md b/AGENTS.md
@@ -234,7 +234,7 @@ src/
 #### Imports & Dependencies
 - Use absolute imports for internal modules: `from authentication import get_auth_dependency`
 - FastAPI dependencies: `from fastapi import APIRouter, HTTPException, Request, status, Depends`
-- Llama Stack imports: `from llama_stack_client import AsyncLlamaStackClient`
+- Llama Stack imports: `from ogx_client import AsyncOgxClient`
 - **ALWAYS** check `pyproject.toml` for existing dependencies before adding new ones
 - **ALWAYS** verify current library versions in `pyproject.toml` rather than assuming versions
 - Check `constants.py` for shared constants before defining new ones

diff --git a/Makefile b/Makefile
@@ -108,7 +108,7 @@ start-llama-stack-container: build-llama-stack-image ## Start llama-stack contai
 		-e WATSONX_API_KEY \
 		-e LITELLM_DROP_PARAMS=true \
 		-e AWS_BEARER_TOKEN_BEDROCK \
-		-e LLAMA_STACK_LOGGING=$${LLAMA_STACK_LOGGING:-} \
+		-e OGX_LOGGING=$${OGX_LOGGING:-} \
 		-e FAISS_VECTOR_STORE_ID=$${FAISS_VECTOR_STORE_ID:-} \
 		-e RH_SERVER_OKP \
 		-e SOLR_URL \
@@ -144,7 +144,7 @@ clean-llama-stack: remove-llama-stack-container ## Remove container and image
 
 run-llama-stack: ## Start Llama Stack with enriched config (for local service mode)
 	uv run src/llama_stack_configuration.py -c $(CONFIG) -i $(LLAMA_STACK_CONFIG) -o $(LLAMA_STACK_CONFIG) && \
-	uv run llama stack run $(LLAMA_STACK_CONFIG)
+	uv run ogx stack run $(LLAMA_STACK_CONFIG)
 
 test-unit: ## Run the unit tests
 	@echo "Running unit tests..."

diff --git a/README.md b/README.md
@@ -770,22 +770,29 @@ For the configuration guide, skill authoring instructions, and examples, see the
 
 ## Safety Shields
 
-A single Llama Stack configuration file can include multiple safety shields, which are utilized in agent
-configurations to monitor input and/or output streams. LCS uses the following naming convention to specify how each safety shield is
-utilized:
+Safety shields used by `/query`, `/streaming_query`, `/responses`, and `/rlsapi`
+are **owned by Lightspeed Core Stack** and configured in `lightspeed-stack.yaml`
+(not via the Llama Stack / OGX Safety or Moderations APIs).
 
-1. If the `shield_id` starts with `input_`, it will be used for input only.
-1. If the `shield_id` starts with `output_`, it will be used for output only.
-1. If the `shield_id` starts with `inout_`, it will be used both for input and output.
-1. Otherwise, it will be used for input only.
+Supported shield types (`provider_id`):
 
-Additionally, an optional list parameter `shield_ids` can be specified in `/query` and `/streaming_query` endpoints to override which shields are applied. You can use this config to disable shield overrides:
+- `question_validity` — topic / off-topic classification
+- `redaction` — regex-based PII redaction
+
+List configured shields with `GET /v1/shields`. Optionally override which
+shields apply with the `shield_ids` request field (`null` = all, `[]` = none,
+or a list of configured `identifier` values). To forbid client overrides on
+`/query` and `/streaming_query`:
 
 ```yaml
 customization:
   disable_shield_ids_override: true
 ```
 
+For configuration details, endpoint application (direct-run vs agent
+capabilities), and examples, see the
+[Safety Shields Guide](docs/user_doc/shields_guide.md).
+
 ## Authentication
 
 See [authentication and authorization](docs/auth.md).

diff --git a/docker-compose-library.yaml b/docker-compose-library.yaml
@@ -56,7 +56,7 @@ services:
       # AWS Bedrock
       - AWS_BEARER_TOKEN_BEDROCK=${AWS_BEARER_TOKEN_BEDROCK:-}
       # Enable debug logging if needed
-      - LLAMA_STACK_LOGGING=${LLAMA_STACK_LOGGING:-}
+      - OGX_LOGGING=${OGX_LOGGING:-}
       # FAISS test and inline RAG config
       - FAISS_VECTOR_STORE_ID=${FAISS_VECTOR_STORE_ID:-}
       # Prevent HuggingFace Hub update checks (HTTP 429 rate-limiting in CI from parallel jobs).

diff --git a/docker-compose.yaml b/docker-compose.yaml
@@ -54,7 +54,7 @@ services:
       # AWS Bedrock
       - AWS_BEARER_TOKEN_BEDROCK=${AWS_BEARER_TOKEN_BEDROCK:-}
       # Enable debug logging if needed
-      - LLAMA_STACK_LOGGING=${LLAMA_STACK_LOGGING:-}
+      - OGX_LOGGING=${OGX_LOGGING:-}
       # FAISS test
       - FAISS_VECTOR_STORE_ID=${FAISS_VECTOR_STORE_ID:-}
       # Prevent HuggingFace Hub update checks (HTTP 429 rate-limiting in CI from parallel jobs).

diff --git a/docs/README.md b/docs/README.md
@@ -19,6 +19,8 @@ See the full documentation at [`../README.md`](../README.md) or browse sub-pages
 
 [Agent skills](https://lightspeed-core.github.io/lightspeed-stack/user_doc/skills_guide.html)
 
+[Safety shields](https://lightspeed-core.github.io/lightspeed-stack/user_doc/shields_guide.html)
+
 [A2A [Agent-to-Agent] Protocol](https://lightspeed-core.github.io/lightspeed-stack/user_doc/a2a_protocol.html)
 
 [RAG configuration guide](https://lightspeed-core.github.io/lightspeed-stack/user_doc/rag_guide.html)

diff --git a/docs/basic_info/overview.md b/docs/basic_info/overview.md
@@ -6,9 +6,11 @@
 
 **Lightspeed Core Stack (LCore)** is an enterprise-grade middleware service that provides a robust layer between client applications and AI Large Language Model (LLM) backends. It adds essential enterprise features such as authentication, authorization, quota management, caching, and observability to LLM interactions.
 
-Current version of LCore is built on **Llama Stack** - open-source framework that provides standardized APIs for building LLM applications. Llama Stack offers a unified interface for models, RAG (vector stores), tools, and safety (shields) across different providers. LCore communicates with Llama Stack to orchestrate all LLM operations.
+Current version of LCore is built on **OGX (Llama Stack)** - open-source framework that provides standardized APIs for building LLM applications. OGX offers a unified interface for models, RAG (vector stores), and tools across different providers. LCore communicates with OGX to orchestrate all LLM operations.
 
-To enhance LLM responses, LCore leverages **RAG (Retrieval-Augmented Generation)**, which retrieves relevant context from vector databases before generating answers. Llama Stack manages the vector stores, and LCore queries them to inject relevant documentation, knowledge bases, or previous conversations into the LLM prompt.
+To enhance LLM responses, LCore leverages **RAG (Retrieval-Augmented Generation)**, which retrieves relevant context from vector databases before generating answers. OGX manages the vector stores, and LCore queries them to inject relevant documentation, knowledge bases, or previous conversations into the LLM prompt.
+
+LCore also provides **safety shields** such as topic validation and PII redaction. These are configured in LCore and applied on request endpoints before or during agent processing.
 
 ### Key Features
 

diff --git a/docs/design/conversation-compaction/conversation-compaction.md b/docs/design/conversation-compaction/conversation-compaction.md
@@ -356,7 +356,7 @@ Example config files go in `examples/`.
 ## Test patterns
 
 - Framework: pytest + pytest-asyncio + pytest-mock. unittest is banned by ruff.
-- Mock Llama Stack client: `mocker.AsyncMock(spec=AsyncLlamaStackClient)`.
+- Mock Llama Stack client: `mocker.AsyncMock(spec=AsyncOgxClient)`.
 - Patch at module level: `mocker.patch("utils.responses.compact_conversation_if_needed", ...)`.
 - Async mocking pattern: see `tests/unit/utils/test_shields.py`.
 - Config validation tests: see `tests/unit/models/config/`.

diff --git a/docs/design/human-in-the-loop/human-in-the-loop.md b/docs/design/human-in-the-loop/human-in-the-loop.md
@@ -564,7 +564,7 @@ Example config files go in `examples/`.
 ### Test patterns
 
 - Framework: pytest + pytest-asyncio + pytest-mock. unittest is banned by ruff.
-- Mock Llama Stack client: `mocker.AsyncMock(spec=AsyncLlamaStackClient)`.
+- Mock Llama Stack client: `mocker.AsyncMock(spec=AsyncOgxClient)`.
 - Patch at module level: `mocker.patch("utils.module.function_name", ...)`.
 - Async mocking pattern: see `tests/unit/utils/test_shields.py`.
 - Config validation tests: see `tests/unit/models/config/`.