address pr feedback

.github/workflows/ghrc-push-main.yml

@@ -6,7 +6,7 @@ env:
 on:
   push:
     branches: [main]
-  pull_request:
+  # pull_request:
 
 jobs:
   build-arm64:

arch/arch_config_schema.yaml

@@ -32,9 +32,9 @@ properties:
             type:
               type: string
               enum:
-                - model_listener
-                - prompt_listener
-                - agent_listener
+                - model
+                - prompt
+                - agent
           required:
             - type
             - name

arch/tools/cli/utils.py

@@ -47,7 +47,6 @@ def convert_legacy_listeners(
         "address": "0.0.0.0",
         "timeout": "30s",
         "model_providers": model_providers or [],
-        "protocol": "openai",
     }
 
     prompt_gateway_listener = {
@@ -56,7 +55,6 @@ def convert_legacy_listeners(
         "port": 10000,
         "address": "0.0.0.0",
         "timeout": "30s",
-        "protocol": "openai",
     }
 
     if isinstance(listeners, dict):

arch/tools/test/test_config_generator.py

@@ -112,7 +112,7 @@ agents:
 
 listeners:
   - name: tmobile
-    type: agent_listener
+    type: agent
     router: arch_agent_v2
     agents:
       - name: simple_tmobile_rag_agent

crates/brightstaff/src/handlers/agent_chat_completions.rs

@@ -129,7 +129,7 @@ async fn handle_agent_chat(
     debug!("Terminal agent details: {:?}", terminal_agent);
 
     let llm_response = pipeline_processor
-        .send_terminal_request(
+        .invoke_upstream_agent(
             &processed_messages,
             &chat_completions_request,
             terminal_agent,

crates/brightstaff/src/handlers/pipeline_processor.rs

@@ -64,7 +64,7 @@ impl PipelineProcessor {
             debug!("Agent details: {:?}", agent);
 
             let response_content = self
-                .send_agent_request(
+                .send_agent_filter_chain_request(
                     &chat_completions_history,
                     initial_request,
                     agent,
@@ -88,7 +88,7 @@ impl PipelineProcessor {
     }
 
     /// Send request to a specific agent and return the response content
-    async fn send_agent_request(
+    async fn send_agent_filter_chain_request(
         &self,
         messages: &[Message],
         original_request: &ChatCompletionsRequest,
@@ -141,7 +141,7 @@ impl PipelineProcessor {
     }
 
     /// Send request to terminal agent and return the raw response for streaming
-    pub async fn send_terminal_request(
+    pub async fn invoke_upstream_agent(
         &self,
         messages: &[Message],
         original_request: &ChatCompletionsRequest,

demos/use_cases/llm_routing/arch_config.yaml

@@ -1,8 +1,8 @@
 version: v0.3.0
 
 listeners:
-  - type: model_listener
-    name: model_listener_1
+  - type: model
+    name: model_1
     address: 0.0.0.0
     port: 12000
 

demos/use_cases/rag_agent/arch_config.yaml

@@ -28,8 +28,8 @@ model_aliases:
     target: gpt-4o
 
 listeners:
-  - type: agent_listener
-    name: agent_listener_1
+  - type: agent
+    name: agent_1
     port: 8001
     router: arch_agent_router
     agents:

docs/source/resources/includes/arch_config_full_reference_rendered.yaml

@@ -31,13 +31,11 @@ listeners:
     provider_interface: mistral
   name: egress_traffic
   port: 12000
-  protocol: openai
   timeout: 5s
   type: model_listener
 - address: 0.0.0.0
   name: ingress_traffic
   port: 10000
-  protocol: openai
   timeout: 5s
   type: prompt_listener
 model_aliases: