Use mcp tools for filter chain (#621)

* agents framework demo

* more changes

* add more changes

* pending changes

* fix tests

* fix more

* rebase with main and better handle error from mcp

* add trace for filters

* add test for client error, server error and for mcp error

* update schema validate code and rename kind => type in agent_filter

* fix agent description and pre-commit

* fix tests

* add provider specific request parsing in agents chat

* fix precommit and tests

* cleanup demo

* update readme

* fix pre-commit

* refactor tracing

* fix fmt

* fix: handle MessageContent enum in responses API conversion

- Update request.rs to handle new MessageContent enum structure from main
- MessageContent can now be Text(String) or Items(Vec<InputContent>)
- Handle new InputItem variants (ItemReference, FunctionCallOutput)
- Fixes compilation error after merging latest main (#632)

* address pr feedback

* fix span

* fix build

* update openai version

arch/arch_config_schema.yaml

@@ -14,6 +14,38 @@ properties:
     type: array
     items:
       type: object
+      properties:
+        id:
+          type: string
+        url:
+          type: string
+      additionalProperties: false
+      required:
+        - id
+        - url
+  filters:
+    type: array
+    items:
+      type: object
+      properties:
+        id:
+          type: string
+        url:
+          type: string
+        type:
+          type: string
+          enum:
+            - mcp
+        transport:
+          type: string
+          enum:
+            - streamable-http
+        tool:
+          type: string
+      additionalProperties: false
+      required:
+        - id
+        - url
   listeners:
     oneOf:
       - type: array

arch/envoy.template.yaml

@@ -214,21 +214,21 @@ static_resources:
             - name: envoy.filters.network.http_connection_manager
               typed_config:
                 "@type": type.googleapis.com/envoy.extensions.filters.network.http_connection_manager.v3.HttpConnectionManager
-                {% if "random_sampling" in arch_tracing and arch_tracing["random_sampling"] > 0 %}
-                generate_request_id: true
-                tracing:
-                  provider:
-                    name: envoy.tracers.opentelemetry
-                    typed_config:
-                      "@type": type.googleapis.com/envoy.config.trace.v3.OpenTelemetryConfig
-                      grpc_service:
-                        envoy_grpc:
-                          cluster_name: opentelemetry_collector
-                        timeout: 0.250s
-                      service_name: tools
-                  random_sampling:
-                    value: {{ arch_tracing.random_sampling }}
-                {% endif %}
+                # {% if "random_sampling" in arch_tracing and arch_tracing["random_sampling"] > 0 %}
+                # generate_request_id: true
+                # tracing:
+                #   provider:
+                #     name: envoy.tracers.opentelemetry
+                #     typed_config:
+                #       "@type": type.googleapis.com/envoy.config.trace.v3.OpenTelemetryConfig
+                #       grpc_service:
+                #         envoy_grpc:
+                #           cluster_name: opentelemetry_collector
+                #         timeout: 0.250s
+                #       service_name: tools
+                #   random_sampling:
+                #     value: {{ arch_tracing.random_sampling }}
+                # {% endif %}
                 stat_prefix: outbound_api_traffic
                 codec_type: AUTO
                 scheme_header_transformation:
@@ -299,7 +299,7 @@ static_resources:
                         envoy_grpc:
                           cluster_name: opentelemetry_collector
                         timeout: 0.250s
-                      service_name: arch_gateway
+                      service_name: plano(inbound)
                   random_sampling:
                     value: {{ arch_tracing.random_sampling }}
                 {% endif %}

arch/tools/cli/config_generator.py

@@ -101,8 +101,17 @@ def validate_and_render_schema():
 
     # Process agents section and convert to endpoints
     agents = config_yaml.get("agents", [])
-    for agent in agents:
+    filters = config_yaml.get("filters", [])
+    agents_combined = agents + filters
+    agent_id_keys = set()
+
+    for agent in agents_combined:
         agent_id = agent.get("id")
+        if agent_id in agent_id_keys:
+            raise Exception(
+                f"Duplicate agent id {agent_id}, please provide unique id for each agent"
+            )
+        agent_id_keys.add(agent_id)
         agent_endpoint = agent.get("url")
 
         if agent_id and agent_endpoint:

arch/tools/cli/utils.py

@@ -57,6 +57,10 @@ def convert_legacy_listeners(
         "timeout": "30s",
     }
 
+    # Handle None case
+    if listeners is None:
+        return [llm_gateway_listener], llm_gateway_listener, prompt_gateway_listener
+
     if isinstance(listeners, dict):
         # legacy listeners
         # check if type is array or object

arch/tools/test/test_config_generator.py

@@ -94,21 +94,16 @@ def test_validate_and_render_happy_path_agent_config(monkeypatch):
 version: v0.3.0
 
 agents:
-  - name: query_rewriter
-    kind: openai
-    endpoint: http://localhost:10500
-  - name: context_builder
-    kind: openai
-    endpoint: http://localhost:10501
-  - name: response_generator
-    kind: openai
-    endpoint: http://localhost:10502
-  - name: research_agent
-    kind: openai
-    endpoint: http://localhost:10500
-  - name: input_guard_rails
-    kind: openai
-    endpoint: http://localhost:10503
+  - id: query_rewriter
+    url: http://localhost:10500
+  - id: context_builder
+    url: http://localhost:10501
+  - id: response_generator
+    url: http://localhost:10502
+  - id: research_agent
+    url: http://localhost:10500
+  - id: input_guard_rails
+    url: http://localhost:10503
 
 listeners:
   - name: tmobile
@@ -156,7 +151,7 @@ listeners:
         mock.mock_open().return_value,  # ARCH_CONFIG_FILE_RENDERED (write)
     ]
     with mock.patch("builtins.open", m_open):
-        with mock.patch("config_generator.Environment"):
+        with mock.patch("cli.config_generator.Environment"):
             validate_and_render_schema()