fix: bypass StreamConsumer — use direct sendFormatted for reliable delivery

src/bot/index.js

@@ -676,13 +676,9 @@ export async function initBot(config, api, tools, skills, agents) {
     await queueRequest(key, text, async () => {
       await ctx.api.sendChatAction(ctx.chat.id, 'typing');
       
-      // Create stream consumer for real-time edit-in-place
-      const consumer = new StreamConsumer(ctx, { editInterval: 1000 });
-      const runPromise = consumer.run();
-      
-      // Wrap chatWithAI with self-correction + streaming
+      // Use self-correction + direct send (streaming disabled for reliability)
       const chatWithCorrection = withSelfCorrection(async (msgs) => {
-        return await chatWithAI(msgs, { onDelta: (token) => consumer.onDelta(token) });
+        return await chatWithAI(msgs);
       });
       
       const result = await chatWithCorrection([
@@ -690,13 +686,13 @@ export async function initBot(config, api, tools, skills, agents) {
         { role: 'user', content: text },
       ]);
       
-      // Signal completion and wait for final edit
-      consumer.finish();
-      await runPromise;
-      
-      // If streaming failed to deliver (no message sent), fallback to plain send
-      if (!consumer.alreadySent && result) {
+      // Send formatted response
+      if (result) {
+        logger.info(`📤 Sending response (${result.length} chars)`);
         await sendFormatted(ctx, result);
+        logger.info('✅ Response delivered');
+      } else {
+        logger.warn('Empty response from AI — nothing to send');
       }
 
       // ── Self-learning: extract patterns from this interaction ──