fix: preload deferred tools by default in yolo mode

CHANGELOG.md

@@ -15,6 +15,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Fixed
 - Model selection and config validation now accept any valid `deepseek-*` model ID (including future releases), while still normalizing common aliases like `deepseek-v3.2` and `deepseek-r1`.
 - Tool-call recovery now auto-loads deferred tools when the model requests them directly, instead of failing with manual `tool_search_*` instructions.
+- YOLO mode now preloads tools by default (including deferred MCP tools), so model tool calls can run immediately without discovery indirection.
 - Unknown tool-call failures now include discovery guidance and nearest tool-name suggestions instead of generic availability errors.
 - Slash-command errors now suggest the closest known command (for example `/modle` -> `/model`) instead of only returning a generic unknown-command message.
 

src/core/engine.rs

@@ -533,7 +533,11 @@ fn is_tool_search_tool(name: &str) -> bool {
     matches!(name, TOOL_SEARCH_REGEX_NAME | TOOL_SEARCH_BM25_NAME)
 }
 
-fn should_default_defer_tool(name: &str) -> bool {
+fn should_default_defer_tool(name: &str, mode: AppMode) -> bool {
+    if mode == AppMode::Yolo {
+        return false;
+    }
+
     !matches!(
         name,
         "read_file"
@@ -1504,13 +1508,17 @@ impl Engine {
         let tools = tool_registry.as_ref().map(|registry| {
             let mut tools = registry.to_api_tools();
             for tool in &mut tools {
-                tool.defer_loading = Some(should_default_defer_tool(&tool.name));
+                tool.defer_loading = Some(should_default_defer_tool(&tool.name, mode));
             }
             let mut mcp_tools = mcp_tools;
             for tool in &mut mcp_tools {
-                let name = tool.name.as_str();
+                if mode == AppMode::Yolo {
+                    tool.defer_loading = Some(false);
+                    continue;
+                }
+
                 let keep_loaded = matches!(
-                    name,
+                    tool.name.as_str(),
                     "list_mcp_resources"
                         | "list_mcp_resource_templates"
                         | "mcp_read_resource"

src/core/engine/tests.rs

@@ -100,6 +100,21 @@ fn tool_exec_outcome_tracks_duration() {
     assert!(outcome.started_at.elapsed().as_nanos() > 0);
 }
 
+#[test]
+fn yolo_mode_keeps_tools_preloaded() {
+    assert!(!should_default_defer_tool("exec_shell", AppMode::Yolo));
+    assert!(!should_default_defer_tool(
+        "mcp_read_resource",
+        AppMode::Yolo
+    ));
+}
+
+#[test]
+fn non_yolo_mode_retains_default_defer_policy() {
+    assert!(should_default_defer_tool("exec_shell", AppMode::Agent));
+    assert!(!should_default_defer_tool("read_file", AppMode::Agent));
+}
+
 #[test]
 fn detects_context_length_errors_from_provider_payloads() {
     let msg = r#"SSE stream request failed: HTTP 400 Bad Request: {"error":{"message":"This model's maximum context length is 131072 tokens. However, you requested 153056 tokens (148960 in the messages, 4096 in the completion).","type":"invalid_request_error"}}"#;