hedgequantx 2.9.1 → 2.9.2

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "hedgequantx",
-  "version": "2.9.1",
+  "version": "2.9.2",
   "description": "HedgeQuantX - Prop Futures Trading CLI",
   "main": "src/app.js",
   "bin": {

package/src/pages/ai-agents.js

@@ -33,12 +33,10 @@ const AI_PROVIDERS = [
   { id: 'qwen', name: 'Qwen', color: 'cyan', supportsOAuth: true, supportsApiKey: true },
   { id: 'iflow', name: 'iFlow (DeepSeek/GLM)', color: 'yellow', supportsOAuth: true, supportsApiKey: true },
   // API Key only (no OAuth - uses LLM Proxy via LiteLLM)
-  { id: 'deepseek', name: 'DeepSeek', color: 'blue', supportsOAuth: false, supportsApiKey: true },
   { id: 'minimax', name: 'MiniMax', color: 'magenta', supportsOAuth: false, supportsApiKey: true },
   { id: 'mistral', name: 'Mistral AI', color: 'yellow', supportsOAuth: false, supportsApiKey: true },
   { id: 'groq', name: 'Groq', color: 'cyan', supportsOAuth: false, supportsApiKey: true },
   { id: 'xai', name: 'xAI (Grok)', color: 'white', supportsOAuth: false, supportsApiKey: true },
-  { id: 'perplexity', name: 'Perplexity', color: 'blue', supportsOAuth: false, supportsApiKey: true },
   { id: 'openrouter', name: 'OpenRouter', color: 'gray', supportsOAuth: false, supportsApiKey: true },
 ];
 

package/src/pages/ai-models.js

@@ -2,28 +2,37 @@
  * AI Models - Fetch from provider APIs
  * 
  * Models are fetched dynamically from each provider's API.
- * No hardcoded model lists - data comes from real APIs only.
+ * Exception: MiniMax (no /models API) - see RULES.md for details.
  */
 
 const https = require('https');
 
 /**
  * API endpoints for fetching models
- * Using beta endpoints where available for latest models
+ * null = provider doesn't have /models endpoint
  */
 const API_ENDPOINTS = {
   anthropic: 'https://api.anthropic.com/v1/models',
   openai: 'https://api.openai.com/v1/models',
-  google: 'https://generativelanguage.googleapis.com/v1beta/models', // v1beta for Gemini 3
-  deepseek: 'https://api.deepseek.com/v1/models',
-  minimax: 'https://api.minimax.chat/v1/models',
+  google: 'https://generativelanguage.googleapis.com/v1beta/models',
+  minimax: null, // No /models API - uses MINIMAX_MODELS (see RULES.md exception)
   mistral: 'https://api.mistral.ai/v1/models',
   groq: 'https://api.groq.com/openai/v1/models',
   xai: 'https://api.x.ai/v1/models',
-  perplexity: 'https://api.perplexity.ai/models',
   openrouter: 'https://openrouter.ai/api/v1/models',
 };
 
+/**
+ * MiniMax Models - EXCEPTION to no-hardcode rule (see RULES.md)
+ * 
+ * MiniMax does not provide /models API endpoint.
+ * Confirmed by: OpenCode, Cursor, LiteLLM - all use hardcoded models.
+ * Source: https://platform.minimax.io/docs/api-reference/text-intro
+ */
+const MINIMAX_MODELS = [
+  { id: 'MiniMax-M2.1', name: 'MiniMax-M2.1' },
+];
+
 /**
  * Make HTTPS request
  * @param {string} url - API URL
@@ -332,7 +341,13 @@ const parseModelsResponse = (providerId, data) => {
  * @returns {Promise<Object>} { success, models, error }
  */
 const fetchModelsFromApi = async (providerId, apiKey) => {
+  // MiniMax: no /models API, use hardcoded list (see RULES.md exception)
+  if (providerId === 'minimax') {
+    return { success: true, models: MINIMAX_MODELS, error: null };
+  }
+  
   const endpoint = API_ENDPOINTS[providerId];
+  
   if (!endpoint) {
     return { success: false, models: [], error: 'Unknown provider' };
   }