🧠 feat: Reasoning UI for Agents (#5904)

* chore: bump https-proxy-agent and @librechat/agents

* refactor: Improve error logging in OllamaClient for API fetch failures

* feat: Add DeepSeek provider support and enhance provider name handling

* refactor: Use Providers.OLLAMA constant for model name check in fetchModels function

* feat: Enhance formatAgentMessages to handle reasoning content type

* feat: OpenRouter Agent Reasoning

* hard work and dedicationgit add .env.example :)

* fix: Handle Google social login with missing last name

Social login with Google was previously displaying 'undefined' when
a user's last name was empty or not provided.

Changes:
- Conditionally render last name only if it exists
- Prevent displaying 'undefined' when last name is missing

* fix: add missing file endings for developers yml,yaml and log

---------

Co-authored-by: Mohamed Al-Duraji <mbalduraji@college.harvard.edu>
Co-authored-by: Deepak Kendole <deepakdpk101@gmail.com>
Co-authored-by: Peter Rothlaender <peter.rothlaender@ginkgo.com>

api/server/controllers/agents/callbacks.js

@@ -199,6 +199,22 @@ function getDefaultHandlers({ res, aggregateContent, toolEndCallback, collectedU
         aggregateContent({ event, data });
       },
     },
+    [GraphEvents.ON_REASONING_DELTA]: {
+      /**
+       * Handle ON_REASONING_DELTA event.
+       * @param {string} event - The event name.
+       * @param {StreamEventData} data - The event data.
+       * @param {GraphRunnableConfig['configurable']} [metadata] The runnable metadata.
+       */
+      handle: (event, data, metadata) => {
+        if (metadata?.last_agent_index === metadata?.agent_index) {
+          sendEvent(res, { event, data });
+        } else if (!metadata?.hide_sequential_outputs) {
+          sendEvent(res, { event, data });
+        }
+        aggregateContent({ event, data });
+      },
+    },
   };
 
   return handlers;

api/server/controllers/agents/client.js

@@ -20,11 +20,6 @@ const {
   bedrockOutputParser,
   removeNullishValues,
 } = require('librechat-data-provider');
-const {
-  extractBaseURL,
-  // constructAzureURL,
-  // genAzureChatCompletion,
-} = require('~/utils');
 const {
   formatMessage,
   formatAgentMessages,
@@ -477,19 +472,6 @@ class AgentClient extends BaseClient {
         abortController = new AbortController();
       }
 
-      const baseURL = extractBaseURL(this.completionsUrl);
-      logger.debug('[api/server/controllers/agents/client.js] chatCompletion', {
-        baseURL,
-        payload,
-      });
-
-      // if (this.useOpenRouter) {
-      //   opts.defaultHeaders = {
-      //     'HTTP-Referer': 'https://librechat.ai',
-      //     'X-Title': 'LibreChat',
-      //   };
-      // }
-
       // if (this.options.headers) {
       //   opts.defaultHeaders = { ...opts.defaultHeaders, ...this.options.headers };
       // }
@@ -626,7 +608,7 @@ class AgentClient extends BaseClient {
         let systemContent = [
           systemMessage,
           agent.instructions ?? '',
-          i !== 0 ? agent.additional_instructions ?? '' : '',
+          i !== 0 ? (agent.additional_instructions ?? '') : '',
         ]
           .join('\n')
           .trim();

api/server/controllers/agents/run.js

@@ -1,5 +1,5 @@
 const { Run, Providers } = require('@librechat/agents');
-const { providerEndpointMap } = require('librechat-data-provider');
+const { providerEndpointMap, KnownEndpoints } = require('librechat-data-provider');
 
 /**
  * @typedef {import('@librechat/agents').t} t
@@ -7,6 +7,7 @@ const { providerEndpointMap } = require('librechat-data-provider');
  * @typedef {import('@librechat/agents').StreamEventData} StreamEventData
  * @typedef {import('@librechat/agents').EventHandler} EventHandler
  * @typedef {import('@librechat/agents').GraphEvents} GraphEvents
+ * @typedef {import('@librechat/agents').LLMConfig} LLMConfig
  * @typedef {import('@librechat/agents').IState} IState
  */
 
@@ -32,6 +33,7 @@ async function createRun({
   streamUsage = true,
 }) {
   const provider = providerEndpointMap[agent.provider] ?? agent.provider;
+  /** @type {LLMConfig} */
   const llmConfig = Object.assign(
     {
       provider,
@@ -41,6 +43,11 @@ async function createRun({
     agent.model_parameters,
   );
 
+  /** @type {'reasoning_content' | 'reasoning'} */
+  let reasoningKey;
+  if (llmConfig.configuration?.baseURL.includes(KnownEndpoints.openrouter)) {
+    reasoningKey = 'reasoning';
+  }
   if (/o1(?!-(?:mini|preview)).*$/.test(llmConfig.model)) {
     llmConfig.streaming = false;
     llmConfig.disableStreaming = true;
@@ -50,6 +57,7 @@ async function createRun({
   const graphConfig = {
     signal,
     llmConfig,
+    reasoningKey,
     tools: agent.tools,
     instructions: agent.instructions,
     additional_instructions: agent.additional_instructions,