chore(cleanup)!: remove tool_runtime.rag_tool (#3871)

Kill the `builtin::rag` tool group completely since it is no longer
targeted. We use the Responses implementation for knowledge_search which
uses the `openai_vector_stores` pathway.

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

llama_stack/distributions/ci-tests/build.yaml

@@ -48,7 +48,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
     - provider_type: remote::model-context-protocol
     batches:
     - provider_type: inline::reference

llama_stack/distributions/ci-tests/run.yaml

@@ -216,8 +216,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
   batches:
@@ -263,8 +261,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/dell/build.yaml

@@ -26,7 +26,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
 image_type: venv
 additional_pip_packages:
 - aiosqlite

llama_stack/distributions/dell/dell.py

@@ -45,7 +45,6 @@ def get_distribution_template() -> DistributionTemplate:
         "tool_runtime": [
             BuildProvider(provider_type="remote::brave-search"),
             BuildProvider(provider_type="remote::tavily-search"),
-            BuildProvider(provider_type="inline::rag-runtime"),
         ],
     }
     name = "dell"
@@ -98,10 +97,6 @@ def get_distribution_template() -> DistributionTemplate:
             toolgroup_id="builtin::websearch",
             provider_id="brave-search",
         ),
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
     ]
 
     return DistributionTemplate(

llama_stack/distributions/dell/run-with-safety.yaml

@@ -87,8 +87,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
 storage:
   backends:
     kv_default:
@@ -133,8 +131,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: brave-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/dell/run.yaml

@@ -83,8 +83,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
 storage:
   backends:
     kv_default:
@@ -124,8 +122,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: brave-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/meta-reference-gpu/build.yaml

@@ -24,7 +24,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
     - provider_type: remote::model-context-protocol
 image_type: venv
 additional_pip_packages:

llama_stack/distributions/meta-reference-gpu/meta_reference.py

@@ -47,7 +47,6 @@ def get_distribution_template() -> DistributionTemplate:
         "tool_runtime": [
             BuildProvider(provider_type="remote::brave-search"),
             BuildProvider(provider_type="remote::tavily-search"),
-            BuildProvider(provider_type="inline::rag-runtime"),
             BuildProvider(provider_type="remote::model-context-protocol"),
         ],
     }
@@ -92,10 +91,6 @@ def get_distribution_template() -> DistributionTemplate:
             toolgroup_id="builtin::websearch",
             provider_id="tavily-search",
         ),
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
     ]
 
     return DistributionTemplate(

llama_stack/distributions/meta-reference-gpu/run-with-safety.yaml

@@ -98,8 +98,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
 storage:
@@ -146,8 +144,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/meta-reference-gpu/run.yaml

@@ -88,8 +88,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
 storage:
@@ -131,8 +129,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/nvidia/build.yaml

@@ -19,8 +19,7 @@ distribution_spec:
     - provider_type: remote::nvidia
     scoring:
     - provider_type: inline::basic
-    tool_runtime:
-    - provider_type: inline::rag-runtime
+    tool_runtime: []
     files:
     - provider_type: inline::localfs
 image_type: venv

llama_stack/distributions/nvidia/nvidia.py

@@ -28,7 +28,7 @@ def get_distribution_template(name: str = "nvidia") -> DistributionTemplate:
             BuildProvider(provider_type="remote::nvidia"),
         ],
         "scoring": [BuildProvider(provider_type="inline::basic")],
-        "tool_runtime": [BuildProvider(provider_type="inline::rag-runtime")],
+        "tool_runtime": [],
         "files": [BuildProvider(provider_type="inline::localfs")],
     }
 
@@ -66,12 +66,7 @@ def get_distribution_template(name: str = "nvidia") -> DistributionTemplate:
         provider_id="nvidia",
     )
 
-    default_tool_groups = [
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
-    ]
+    default_tool_groups: list[ToolGroupInput] = []
 
     return DistributionTemplate(
         name=name,

llama_stack/distributions/nvidia/run-with-safety.yaml

@@ -80,9 +80,7 @@ providers:
   scoring:
   - provider_id: basic
     provider_type: inline::basic
-  tool_runtime:
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
+  tool_runtime: []
   files:
   - provider_id: meta-reference-files
     provider_type: inline::localfs
@@ -128,9 +126,7 @@ registered_resources:
   datasets: []
   scoring_fns: []
   benchmarks: []
-  tool_groups:
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
+  tool_groups: []
 server:
   port: 8321
 telemetry:

llama_stack/distributions/nvidia/run.yaml

@@ -69,9 +69,7 @@ providers:
   scoring:
   - provider_id: basic
     provider_type: inline::basic
-  tool_runtime:
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
+  tool_runtime: []
   files:
   - provider_id: meta-reference-files
     provider_type: inline::localfs
@@ -107,9 +105,7 @@ registered_resources:
   datasets: []
   scoring_fns: []
   benchmarks: []
-  tool_groups:
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
+  tool_groups: []
 server:
   port: 8321
 telemetry:

llama_stack/distributions/open-benchmark/build.yaml

@@ -28,7 +28,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
     - provider_type: remote::model-context-protocol
 image_type: venv
 additional_pip_packages:

llama_stack/distributions/open-benchmark/open_benchmark.py

@@ -118,7 +118,6 @@ def get_distribution_template() -> DistributionTemplate:
         "tool_runtime": [
             BuildProvider(provider_type="remote::brave-search"),
             BuildProvider(provider_type="remote::tavily-search"),
-            BuildProvider(provider_type="inline::rag-runtime"),
             BuildProvider(provider_type="remote::model-context-protocol"),
         ],
     }
@@ -154,10 +153,6 @@ def get_distribution_template() -> DistributionTemplate:
             toolgroup_id="builtin::websearch",
             provider_id="tavily-search",
         ),
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
     ]
 
     models, _ = get_model_registry(available_models)

llama_stack/distributions/open-benchmark/run.yaml

@@ -118,8 +118,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
 storage:
@@ -244,8 +242,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/postgres-demo/build.yaml

@@ -14,7 +14,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
     - provider_type: remote::model-context-protocol
 image_type: venv
 additional_pip_packages:

llama_stack/distributions/postgres-demo/postgres_demo.py

@@ -45,7 +45,6 @@ def get_distribution_template() -> DistributionTemplate:
         "tool_runtime": [
             BuildProvider(provider_type="remote::brave-search"),
             BuildProvider(provider_type="remote::tavily-search"),
-            BuildProvider(provider_type="inline::rag-runtime"),
             BuildProvider(provider_type="remote::model-context-protocol"),
         ],
     }
@@ -66,10 +65,6 @@ def get_distribution_template() -> DistributionTemplate:
             toolgroup_id="builtin::websearch",
             provider_id="tavily-search",
         ),
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
     ]
 
     default_models = [

llama_stack/distributions/postgres-demo/run.yaml

@@ -54,8 +54,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
 storage:
@@ -107,8 +105,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/starter-gpu/build.yaml

@@ -49,7 +49,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
     - provider_type: remote::model-context-protocol
     batches:
     - provider_type: inline::reference

llama_stack/distributions/starter-gpu/run.yaml

@@ -219,8 +219,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
   batches:
@@ -266,8 +264,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/starter/build.yaml

@@ -49,7 +49,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
     - provider_type: remote::model-context-protocol
     batches:
     - provider_type: inline::reference

llama_stack/distributions/starter/run.yaml

@@ -216,8 +216,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
   batches:
@@ -263,8 +261,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/starter/starter.py

@@ -140,7 +140,6 @@ def get_distribution_template(name: str = "starter") -> DistributionTemplate:
         "tool_runtime": [
             BuildProvider(provider_type="remote::brave-search"),
             BuildProvider(provider_type="remote::tavily-search"),
-            BuildProvider(provider_type="inline::rag-runtime"),
             BuildProvider(provider_type="remote::model-context-protocol"),
         ],
         "batches": [
@@ -162,10 +161,6 @@ def get_distribution_template(name: str = "starter") -> DistributionTemplate:
             toolgroup_id="builtin::websearch",
             provider_id="tavily-search",
         ),
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
     ]
     default_shields = [
         # if the

llama_stack/distributions/watsonx/build.yaml

@@ -23,7 +23,6 @@ distribution_spec:
     tool_runtime:
     - provider_type: remote::brave-search
     - provider_type: remote::tavily-search
-    - provider_type: inline::rag-runtime
     - provider_type: remote::model-context-protocol
     files:
     - provider_type: inline::localfs

llama_stack/distributions/watsonx/run.yaml

@@ -83,8 +83,6 @@ providers:
     config:
       api_key: ${env.TAVILY_SEARCH_API_KEY:=}
       max_results: 3
-  - provider_id: rag-runtime
-    provider_type: inline::rag-runtime
   - provider_id: model-context-protocol
     provider_type: remote::model-context-protocol
   files:
@@ -125,8 +123,6 @@ registered_resources:
   tool_groups:
   - toolgroup_id: builtin::websearch
     provider_id: tavily-search
-  - toolgroup_id: builtin::rag
-    provider_id: rag-runtime
 server:
   port: 8321
 telemetry:

llama_stack/distributions/watsonx/watsonx.py

@@ -33,7 +33,6 @@ def get_distribution_template(name: str = "watsonx") -> DistributionTemplate:
         "tool_runtime": [
             BuildProvider(provider_type="remote::brave-search"),
             BuildProvider(provider_type="remote::tavily-search"),
-            BuildProvider(provider_type="inline::rag-runtime"),
             BuildProvider(provider_type="remote::model-context-protocol"),
         ],
         "files": [BuildProvider(provider_type="inline::localfs")],
@@ -50,10 +49,6 @@ def get_distribution_template(name: str = "watsonx") -> DistributionTemplate:
             toolgroup_id="builtin::websearch",
             provider_id="tavily-search",
         ),
-        ToolGroupInput(
-            toolgroup_id="builtin::rag",
-            provider_id="rag-runtime",
-        ),
     ]
 
     files_provider = Provider(