Bump version to 2.0.0 and upload install-galaxy.sh in deploy script

- Update version from 1.6.3 to 2.0.0 in app/Cargo.toml and Cargo.lock
- Add install-galaxy.sh upload step to build-and-deploy-hermes script
- Include pending AI provider and agent changes

app/src/ai/openai/client.rs

@@ -9,6 +9,8 @@ pub struct OpenAIClientConfig {
     pub base_url: String,
     pub api_key: Option<String>,
     pub model: Option<String>,
+    pub max_input_tokens: Option<u32>,
+    pub max_output_tokens: Option<u32>,
 }
 
 pub struct OpenAIClient {

app/src/ai/openai/response_translator.rs

@@ -30,6 +30,7 @@ pub fn openai_stream_to_response_events(
     user_query: Option<String>,
     messages_sent: Arc<Mutex<Vec<ConversationMessage>>>,
     model_id: String,
+    max_context_tokens: Option<u32>,
     _tool_result_archive: Vec<ConversationMessage>,
 ) -> BoxStream<'static, Event> {
     use futures::StreamExt;
@@ -269,7 +270,14 @@ pub fn openai_stream_to_response_events(
         }
 
         let cost = estimate_cost_cents(input_tokens as u32, output_tokens as u32, &model_id);
-        let finished_event = build_stream_finished(stop_reason, input_tokens, output_tokens, cost, &model_id);
+        let finished_event = build_stream_finished(
+            stop_reason,
+            input_tokens,
+            output_tokens,
+            cost,
+            &model_id,
+            max_context_tokens,
+        );
         yield Ok(finished_event);
 
         log::info!("[openai] Stream finished: input_tokens={input_tokens}, output_tokens={output_tokens}");
@@ -409,6 +417,7 @@ fn build_stream_finished(
     output_tokens: i32,
     cost_in_cents: f32,
     model_id: &str,
+    max_context_tokens: Option<u32>,
 ) -> ResponseEvent {
     let total_tokens = (input_tokens + output_tokens) as u32;
 
@@ -434,12 +443,14 @@ fn build_stream_finished(
         cost_in_cents,
     }];
 
-    let max_context_tokens = context_window_for_model(model_id);
+    let max_context_tokens =
+        max_context_tokens.unwrap_or_else(|| context_window_for_model(model_id));
     let context_usage = if max_context_tokens > 0 {
         input_tokens as f32 / max_context_tokens as f32
     } else {
         0.0
-    };
+    }
+    .clamp(0.0, 1.0);
 
     #[allow(deprecated)]
     let conversation_usage_metadata = Some(stream_finished::ConversationUsageMetadata {

app/src/ai/openai/translator.rs

@@ -10,6 +10,8 @@ use crate::ai::agent::api::ResponseStream;
 use crate::ai::bedrock::request_translator;
 use crate::ai::provider::types::{ConversationMessage, MessageContent, MessageRole};
 
+const DEFAULT_MAX_OUTPUT_TOKENS: u32 = 64_000;
+
 pub struct TranslatorRequest {
     pub config: OpenAIClientConfig,
     pub model_id: String,
@@ -98,6 +100,10 @@ pub async fn execute(
         messages.extend(new_input_messages);
     }
 
+    for message in &mut messages {
+        message.truncate_tool_results_for_provider_request();
+    }
+
     sanitize_messages_for_openai(&mut messages);
 
     let system_prompt = request_translator::extract_system_prompt(request);
@@ -112,11 +118,17 @@ pub async fn execute(
 
     let user_query_text = request_translator::extract_user_query_text(request);
 
+    let max_output_tokens = params
+        .config
+        .max_output_tokens
+        .unwrap_or(DEFAULT_MAX_OUTPUT_TOKENS)
+        .min(i32::MAX as u32) as i32;
+
     let request_body = build_openai_request(
         messages.clone(),
         system_prompt,
         tools,
-        64000,
+        max_output_tokens,
         None,
         &model_id,
     );
@@ -140,6 +152,7 @@ pub async fn execute(
         user_query_text,
         params.messages_sent.clone(),
         model_id,
+        params.config.max_input_tokens,
         params.tool_result_archive,
     );