polydev-ai 1.8.13 → 1.8.14

package/lib/cliManager.js

@@ -183,6 +183,7 @@ class CLIManager {
       }
 
       let authenticated = false;
+      let quotaExhausted = false;
       
       // For Claude Code, skip command-based auth check and use file-based detection directly
       // This avoids the recursion issue when running from within Claude Code
@@ -211,6 +212,12 @@ class CLIManager {
           
           authenticated = this.parseAuthenticationStatus(provider.id, authOutput);
           
+          // Check for quota exhaustion (authenticated but rate limited)
+          if (provider.id === 'gemini_cli') {
+            quotaExhausted = authOutput.includes('exhausted your daily quota') ||
+                            authOutput.includes('quota') && authOutput.includes('exhausted');
+          }
+          
         } catch (authError) {
           if (process.env.POLYDEV_CLI_DEBUG) {
             console.log(`[CLI Debug] Auth check failed for ${provider.id}:`, authError);
@@ -262,6 +269,7 @@ This is a known issue with @google/gemini-cli@0.3.4 and older Node.js versions.`
       return {
         available: true,
         authenticated,
+        quota_exhausted: quotaExhausted,
         version,
         path: cliPath,
         last_checked: new Date(),
@@ -437,6 +445,19 @@ This is a known issue with @google/gemini-cli@0.3.4 and older Node.js versions.`
         };
       }
 
+      // Check if CLI has quota exhausted - skip and suggest API fallback
+      if (providerStatus.quota_exhausted) {
+        console.log(`[Polydev CLI] ${provider.name} has exhausted daily quota, skipping CLI (use API fallback)`);
+        return {
+          success: false,
+          error: `${provider.name} has exhausted daily quota. Use API fallback.`,
+          error_code: 'QUOTA_EXHAUSTED',
+          latency_ms: Date.now() - startTime,
+          provider: providerId,
+          timestamp: new Date()
+        };
+      }
+
       // Log model being used
       if (model) {
         console.log(`[Polydev CLI] Using model for ${providerId}: ${model}`);
@@ -611,6 +632,22 @@ This is a known issue with @google/gemini-cli@0.3.4 and older Node.js versions.`
           }
 
           lastErrorMessage = result.error;
+          
+          // Check for quota exhaustion error during execution
+          const combinedOutput = ((result.stdout || '') + ' ' + (result.stderr || '') + ' ' + (result.error || '')).toLowerCase();
+          if (combinedOutput.includes('exhausted') && combinedOutput.includes('quota') ||
+              combinedOutput.includes('rate limit') ||
+              combinedOutput.includes('too many requests')) {
+            console.log(`[Polydev CLI] ${providerId} quota/rate limit error detected during execution`);
+            return {
+              success: false,
+              error: `${provider.name} quota/rate limit exceeded. Use API fallback.`,
+              error_code: 'QUOTA_EXHAUSTED',
+              latency_ms: Date.now() - startTime,
+              provider: providerId,
+              timestamp: new Date()
+            };
+          }
         } catch (error) {
           lastErrorMessage = error instanceof Error ? error.message : String(error);
           

package/mcp/stdio-wrapper.js

@@ -559,6 +559,11 @@ class StdioMCPWrapper {
       for (const providerId of priorityOrder) {
         const status = results[providerId];
         if (status && status.available && status.authenticated) {
+          // Skip providers with quota exhausted - they'll use API fallback
+          if (status.quota_exhausted) {
+            console.error(`[Stdio Wrapper] Skipping ${providerId} - quota exhausted, will use API fallback`);
+            continue;
+          }
           availableProviders.push(providerId);
         }
       }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "polydev-ai",
-  "version": "1.8.13",
+  "version": "1.8.14",
   "description": "Agentic workflow assistant with CLI integration - get diverse perspectives from multiple LLMs when stuck or need enhanced reasoning",
   "keywords": [
     "mcp",