lynkr 9.12.0 → 9.14.0

package/README.md

@@ -15,7 +15,7 @@
 <td align="center"><strong>84%</strong><br/>JSON Compression</td>
 <td align="center"><strong>53%</strong><br/>Tool Token Reduction</td>
 <td align="center"><strong>&lt;300ms</strong><br/>Semantic Cache Hits</td>
-<td align="center"><strong>13+</strong><br/>LLM Providers</td>
+<td align="center"><strong>14+</strong><br/>LLM Providers</td>
 <td align="center"><strong>0</strong><br/>Code Changes Required</td>
 </tr>
 </table>
@@ -28,26 +28,25 @@
 
 ## 🚀 New: Wrap Mode for AI Coding Tools
 
-**Use Lynkr's routing with your AI coding assistant — BYO API keys, route by task complexity:**
+**Use Lynkr's routing with your AI coding assistant — maximize your subscription value:**
 
 ```bash
 npm install -g lynkr
 
 
-# Claude Code (with your own ANTHROPIC_API_KEY)
+# Claude Code Pro/Max
 lynkr wrap claude
 
 
 ```
 
 **Wrapping gives you:**
-- ✅ Tier routing (send simple tasks to free Ollama, complex to your configured API provider)
+- ✅ Tier routing (send simple tasks to free Ollama, complex to your subscription/API)
 - ✅ Sticky sessions: one routing decision per conversation via content fingerprinting, with automatic escalation when the task outgrows the model
 - ✅ TOON/RTK compression (84% token reduction on large JSON tool outputs)
 - ✅ Semantic caching (sub-300ms cache hits, 0 tokens billed)
-- ✅ Works with API keys (Claude, Aider, Codex, Copilot, Cursor)
-
-> **Note on subscription (OAuth) logins:** Anthropic's Feb 2026 usage policy prohibits third-party tools from using OAuth tokens issued to Free/Pro/Max accounts. Wrap mode's tier routing, compression, and caching require an API key — they do not apply to and are not compatible with Claude Pro/Max subscription auth. See [ToS Compliance](docs/wrap-guide.md#tos-compliance) for details.
+- ✅ **3-5x more usage from the same subscription limits**
+- ✅ Works with OAuth (Claude, Copilot, Cursor) or API keys (Aider, Codex)
 
 [Full wrap guide →](docs/wrap-guide.md)
 
@@ -69,7 +68,7 @@ The fastest path is the interactive wizard:
 lynkr init
 ```
 
-It asks four questions — usage mode (BYO API keys, direct), tier picks for SIMPLE/MEDIUM/COMPLEX/REASONING across the 12 supported providers, credentials for what you chose, and a few routing-intelligence knobs — then writes a fully-populated `.env` with sensible production defaults for everything else (caching, compression, policy budgets, MCP sandbox, rate limiting).
+It asks four questions — usage mode (Claude Pro/Max via wrap, or direct API keys), tier picks for SIMPLE/MEDIUM/COMPLEX/REASONING across the 14 supported providers, credentials for what you chose, and a few routing-intelligence knobs — then writes a fully-populated `.env` with sensible production defaults for everything else (caching, compression, policy budgets, MCP sandbox, rate limiting).
 
 Useful flags:
 
@@ -217,9 +216,24 @@ Claude Code / Cursor / Codex / Cline / Continue
 | **Azure OpenAI** | Cloud | GPT-4o, o1, o3 | $$$ |
 | **Azure Anthropic** | Cloud | Claude Sonnet, Opus | $$$ |
 | **OpenAI** | Cloud | GPT-4o, o3-mini | $$$ |
+| **Atlas Cloud** | Cloud | Qwen, DeepSeek, and other OpenAI-compatible models | $-$$$ |
 | **DeepSeek** | Cloud | DeepSeek R1, Reasoner | $ |
-
-**4 local providers** for 100% offline, free usage. **10+ cloud providers** for scale.
+| **Z.ai** | Cloud | GLM-4.7, GLM-4.5-Air | $ |
+| **Moonshot AI** | Cloud | Kimi K2.6, Kimi K3 | $ |
+| **Baidu Qianfan** | Cloud | ERNIE 4.5 Turbo, ERNIE X1.1 | $ (unverified — not yet probed against a live key) |
+
+**4 local providers** for 100% offline, free usage. **14+ cloud providers** for scale.
+
+<div align="center">
+  <a href="https://www.atlascloud.ai/?utm_source=github&amp;utm_medium=readme&amp;utm_campaign=lynkr">
+    <picture>
+      <source media="(prefers-color-scheme: dark)" srcset="https://www.atlascloud.ai/logo-white.svg">
+      <img src="https://www.atlascloud.ai/logo.svg" alt="Atlas Cloud" width="196">
+    </picture>
+  </a>
+  <br>
+  <sub>Use Lynkr with Atlas Cloud's OpenAI-compatible API and route across its live model catalog with one optional provider configuration. <a href="https://www.atlascloud.ai/?utm_source=github&amp;utm_medium=readme&amp;utm_campaign=lynkr">Explore Atlas Cloud</a>.</sub>
+</div>
 
 ---
 

package/bin/lynkr-init.js

@@ -89,6 +89,17 @@ const PROVIDERS = {
     extras: [],
     defaultModel: 'gpt-4o',
   },
+  atlas: {
+    label: 'Atlas Cloud (OpenAI-compatible)',
+    local: false,
+    creds: [
+      { key: 'ATLASCLOUD_API_KEY', label: 'Atlas Cloud API key', secret: true },
+    ],
+    extras: [
+      { key: 'ATLASCLOUD_ENDPOINT', label: 'Chat Completions endpoint', default: 'https://api.atlascloud.ai/v1/chat/completions' },
+    ],
+    defaultModel: 'qwen/qwen3.8-max',
+  },
   openrouter: {
     label: 'OpenRouter (100+ models, one key)',
     local: false,
@@ -159,12 +170,21 @@ const PROVIDERS = {
     extras: [],
     defaultModel: 'kimi-k2-turbo-preview',
   },
+  baidu: {
+    label: 'Baidu Qianfan (ERNIE)',
+    local: false,
+    creds: [
+      { key: 'BAIDU_API_KEY', label: 'Baidu Qianfan API key (bce-v3/ALTAK-...)', secret: true },
+    ],
+    extras: [],
+    defaultModel: 'ernie-4.5-turbo-128k',
+  },
 };
 
 const PROVIDER_ORDER = [
   'ollama', 'llamacpp', 'lmstudio',
-  'azure-anthropic', 'azure-openai', 'openai', 'openrouter', 'edenai',
-  'databricks', 'bedrock', 'vertex', 'zai', 'moonshot',
+  'azure-anthropic', 'azure-openai', 'openai', 'atlas', 'openrouter', 'edenai',
+  'databricks', 'bedrock', 'vertex', 'zai', 'moonshot', 'baidu',
 ];
 const TIERS = ['SIMPLE', 'MEDIUM', 'COMPLEX', 'REASONING'];
 
@@ -319,6 +339,9 @@ const BASELINE_ENV = {
   MOONSHOT_API_KEY: '',
   MOONSHOT_ENDPOINT: 'https://api.moonshot.ai/v1/chat/completions',
   MOONSHOT_MODEL: 'kimi-k2.6',
+  BAIDU_API_KEY: '',
+  BAIDU_ENDPOINT: 'https://qianfan.baidubce.com/v2/chat/completions',
+  BAIDU_MODEL: 'ernie-4.5-turbo-128k',
   LLAMACPP_ENDPOINT: 'http://localhost:8080',
   LLAMACPP_MODEL: 'default',
   LLAMACPP_TIMEOUT_MS: '120000',

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "lynkr",
-  "version": "9.12.0",
+  "version": "9.14.0",
   "files": [
     "index.js",
     "install.sh",
@@ -35,9 +35,10 @@
     "start": "node index.js 2>&1 | npx pino-pretty --sync",
     "stop": "node -e \"if(process.env.HEADROOM_ENABLED==='true'&&process.env.HEADROOM_DOCKER_ENABLED!=='false'){process.exit(0)}else{process.exit(1)}\" && docker compose --profile headroom down || echo 'Headroom skipped (disabled or Docker not running)'",
     "dev": "nodemon index.js",
+    "start:supervised": "while true; do node index.js 2>&1 | npx pino-pretty --sync; code=$?; echo \"[supervisor] lynkr exited ($code) — restarting in 3s\"; sleep 3; done",
     "lint": "eslint src index.js",
     "test": "npm run test:unit && npm run test:performance",
-    "test:unit": "DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com LOG_FILE_ENABLED=false node --test test/routing.test.js test/hybrid-routing-integration.test.js test/retry-logic.test.js test/sse-transformer.test.js test/passthrough-stream.test.js test/passthrough-mode.test.js test/openrouter-error-resilience.test.js test/format-conversion.test.js test/azure-openai-config.test.js test/azure-openai-format-conversion.test.js test/azure-openai-routing.test.js test/azure-openai-streaming.test.js test/azure-openai-error-resilience.test.js test/azure-openai-integration.test.js test/openai-integration.test.js test/toon-compression.test.js test/gcf-compression.test.js test/llamacpp-integration.test.js test/resilience.test.js test/telemetry-routing.test.js test/memory/store.test.js test/memory/surprise.test.js test/memory/extractor.test.js test/memory/search.test.js test/memory/retriever.test.js test/memory/distiller.test.js test/memory/distiller-freeze.test.js test/memory/wiki.test.js test/memory/skills-cache.test.js test/memory/tencentdb-launcher.test.js test/distill.test.js test/large-payload.test.js test/prompt-cache-injection.test.js test/risk-analyzer.test.js test/interaction-block.test.js test/preflight.test.js test/token-reduction.test.js test/session-affinity.test.js test/cache-state.test.js test/cache-switch-cost.test.js test/lens-recommendations.test.js test/model-registry-cost.test.js test/output-format-guard.test.js test/tier-fallback.test.js test/wrap.test.js test/init.test.js test/tool-call-response-metadata.test.js test/degradation.test.js test/routing-telemetry-columns.test.js test/sticky-routing.test.js test/knn-ambiguous-escalate.test.js test/deescalator.test.js test/client-profiles.test.js test/strip-internal-fields.test.js test/complexity-tool-subtraction.test.js test/bandit.test.js test/routing-propensity.test.js test/reward-pipeline.test.js test/knn-cold-start.test.js test/calibration.test.js test/feedback-loop.test.js test/session-fingerprint.test.js test/side-request-guards.test.js test/verifier.test.js test/intent-score.test.js test/difficulty-classifier.test.js test/classifier-setup.test.js test/usage-stats.test.js test/loop-guard.test.js test/moonshot-model-mapping.test.js",
+    "test:unit": "DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com LOG_FILE_ENABLED=false node --test test/routing.test.js test/hybrid-routing-integration.test.js test/retry-logic.test.js test/sse-transformer.test.js test/passthrough-stream.test.js test/passthrough-mode.test.js test/openrouter-error-resilience.test.js test/format-conversion.test.js test/azure-openai-config.test.js test/azure-openai-format-conversion.test.js test/azure-openai-routing.test.js test/azure-openai-streaming.test.js test/azure-openai-error-resilience.test.js test/azure-openai-integration.test.js test/openai-integration.test.js test/atlas-integration.test.js test/toon-compression.test.js test/gcf-compression.test.js test/llamacpp-integration.test.js test/resilience.test.js test/telemetry-routing.test.js test/memory/store.test.js test/memory/surprise.test.js test/memory/extractor.test.js test/memory/search.test.js test/memory/retriever.test.js test/memory/distiller.test.js test/memory/distiller-freeze.test.js test/memory/wiki.test.js test/memory/skills-cache.test.js test/memory/tencentdb-launcher.test.js test/distill.test.js test/large-payload.test.js test/prompt-cache-injection.test.js test/risk-analyzer.test.js test/interaction-block.test.js test/preflight.test.js test/token-reduction.test.js test/session-affinity.test.js test/cache-state.test.js test/cache-switch-cost.test.js test/lens-recommendations.test.js test/model-registry-cost.test.js test/output-format-guard.test.js test/tier-fallback.test.js test/wrap.test.js test/init.test.js test/tool-call-response-metadata.test.js test/degradation.test.js test/routing-telemetry-columns.test.js test/sticky-routing.test.js test/knn-ambiguous-escalate.test.js test/deescalator.test.js test/client-profiles.test.js test/strip-internal-fields.test.js test/complexity-tool-subtraction.test.js test/bandit.test.js test/routing-propensity.test.js test/reward-pipeline.test.js test/knn-cold-start.test.js test/calibration.test.js test/feedback-loop.test.js test/session-fingerprint.test.js test/side-request-guards.test.js test/verifier.test.js test/intent-score.test.js test/difficulty-classifier.test.js test/classifier-setup.test.js test/usage-stats.test.js test/loop-guard.test.js test/moonshot-model-mapping.test.js test/baidu-model-mapping.test.js",
     "test:memory": "DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node --test test/memory/store.test.js test/memory/surprise.test.js test/memory/extractor.test.js test/memory/search.test.js test/memory/retriever.test.js test/memory/distiller.test.js test/memory/distiller-freeze.test.js test/memory/wiki.test.js test/memory/skills-cache.test.js test/memory/tencentdb-launcher.test.js",
     "test:new-features": "DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node --test test/passthrough-mode.test.js test/openrouter-error-resilience.test.js test/format-conversion.test.js",
     "test:performance": "DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node test/hybrid-routing-performance.test.js && DATABRICKS_API_KEY=test-key DATABRICKS_API_BASE=http://test.com node test/performance-tests.js",

package/src/api/middleware/load-shedding.js

@@ -1,4 +1,5 @@
 const os = require("os");
+const v8 = require("v8");
 const logger = require("../../logger");
 const { ServiceUnavailableError } = require("./error-handling");
 
@@ -18,11 +19,31 @@ class LoadShedder {
     this.memoryThreshold = options.memoryThreshold || 0.85; // 85%
     this.heapThreshold = options.heapThreshold || 0.95; // 95% (increased from 90% to prevent false positives from temporary allocation spikes)
     this.activeRequestsThreshold = options.activeRequestsThreshold || 1000;
+    // Absolute free-heap floor. Percent-of-heapTotal is unreliable in both
+    // directions: heapTotal is V8's CURRENTLY allocated heap (grows on
+    // demand toward the real limit), and at the limit "99%" leaves ~40MB on
+    // a 4GB heap — nowhere near enough to parse one multi-MB request body.
+    // 2026-08-19 OOM: a 5MB compact request passed the percent check and
+    // died allocating during JSON.parse. Shed while there is still room to
+    // finish in-flight work.
+    this.headroomMB = options.headroomMB || 512;
+    this.heapLimitMB = v8.getHeapStatistics().heap_size_limit / (1024 * 1024);
+
+    // Watchdog: how long the heap may stay below the headroom floor with
+    // zero active requests before the process self-restarts. Retained
+    // memory GC can't reclaim never recovers on its own — without this the
+    // server sheds 100% of traffic until someone restarts it by hand.
+    this.wedgedRestartMs = options.wedgedRestartMs || 60000;
+    this.selfRestart = options.selfRestart !== false;
+    this.wedgedSince = null;
 
     // State
     this.activeRequests = 0;
     this.totalShed = 0;
-    this.lastCheck = Date.now();
+    // 0, not Date.now(): the first isOverloaded() call must actually
+    // evaluate instead of returning the constructor's cached "false"
+    // for the first checkInterval window.
+    this.lastCheck = 0;
     this.checkInterval = options.checkInterval || 1000; // Check every second
     this.cachedOverloadState = false;
   }
@@ -42,26 +63,32 @@ class LoadShedder {
 
     // Check memory usage
     const memUsage = process.memoryUsage();
-    const heapUsedPercent = memUsage.heapUsed / memUsage.heapTotal;
-
-    // FIX: Only trigger if BOTH percentage is high AND actual usage is significant
-    // This prevents false positives on startup when heapTotal is small but will grow
-    const heapUsedMB = memUsage.heapUsed / (1024 * 1024);
-    const minHeapThresholdMB = 500; // Only shed load if using more than 500MB
 
-    if (heapUsedPercent > this.heapThreshold && heapUsedMB > minHeapThresholdMB) {
+    // Absolute headroom against the true V8 limit — the primary guard.
+    const heapStats = v8.getHeapStatistics();
+    const freeMB = (heapStats.heap_size_limit - heapStats.used_heap_size) / (1024 * 1024);
+    if (freeMB < this.headroomMB) {
       logger.warn(
         {
-          heapUsedPercent: (heapUsedPercent * 100).toFixed(2),
-          heapUsedMB: heapUsedMB.toFixed(2),
-          threshold: (this.heapThreshold * 100).toFixed(2),
-          minThresholdMB: minHeapThresholdMB,
+          freeMB: freeMB.toFixed(0),
+          headroomMB: this.headroomMB,
+          heapLimitMB: this.heapLimitMB.toFixed(0),
+          usedMB: (heapStats.used_heap_size / (1024 * 1024)).toFixed(0),
         },
-        "Load shedding: Heap usage exceeded threshold"
+        "Load shedding: heap headroom below floor"
       );
+      this._trackWedged(now);
       this.cachedOverloadState = true;
       return true;
     }
+    this.wedgedSince = null;
+
+    // NOTE: the old heapUsed/heapTotal percent check is gone. heapTotal is
+    // V8's *currently allocated* heap, which it keeps close to usage, so the
+    // ratio reads ~99% during perfectly normal operation — it shed all
+    // traffic at 755MB used with gigabytes of real headroom remaining
+    // (2026-08-20 false-positive incident). The absolute-headroom check
+    // against heap_size_limit above is the correct form of this guard.
 
     // Check RSS / system memory
     const rssPercent = memUsage.rss / os.totalmem();
@@ -94,6 +121,32 @@ class LoadShedder {
     return false;
   }
 
+  /**
+   * Track how long the heap has been wedged below the headroom floor.
+   * If it persists past wedgedRestartMs with no in-flight requests, the
+   * memory is retained (not transient) and GC cannot recover it — exit so
+   * the supervisor restarts a clean process instead of shedding forever.
+   */
+  _trackWedged(now) {
+    if (!this.selfRestart) return;
+    if (this.wedgedSince === null) {
+      this.wedgedSince = now;
+      return;
+    }
+    if (now - this.wedgedSince >= this.wedgedRestartMs && this.activeRequests === 0) {
+      logger.fatal(
+        {
+          wedgedForMs: now - this.wedgedSince,
+          totalShed: this.totalShed,
+          heapLimitMB: this.heapLimitMB.toFixed(0),
+        },
+        "Load shedding: heap wedged below headroom floor with no active requests — exiting for supervisor restart"
+      );
+      // Give the fatal log a moment to flush, then exit non-zero.
+      setTimeout(() => process.exit(1), 250).unref();
+    }
+  }
+
   /**
    * Get current metrics
    */
@@ -107,10 +160,15 @@ class LoadShedder {
       heapTotalMB: (memUsage.heapTotal / (1024 * 1024)).toFixed(2),
       rssMB: (memUsage.rss / (1024 * 1024)).toFixed(2),
       rssPercent: ((memUsage.rss / os.totalmem()) * 100).toFixed(2),
+      heapLimitMB: this.heapLimitMB.toFixed(2),
+      freeHeadroomMB: ((v8.getHeapStatistics().heap_size_limit - v8.getHeapStatistics().used_heap_size) / (1024 * 1024)).toFixed(2),
       thresholds: {
         heapThreshold: (this.heapThreshold * 100).toFixed(2),
         memoryThreshold: (this.memoryThreshold * 100).toFixed(2),
         activeRequestsThreshold: this.activeRequestsThreshold,
+        headroomMB: this.headroomMB,
+        wedgedRestartMs: this.wedgedRestartMs,
+        selfRestart: this.selfRestart,
       },
     };
   }
@@ -129,6 +187,11 @@ function getLoadShedder(options) {
         process.env.LOAD_SHEDDING_ACTIVE_REQUESTS_THRESHOLD || "1000",
         10
       ),
+      // Not env-configurable by design: these are crash-prevention floors,
+      // and there is no deployment where turning them off is correct.
+      headroomMB: 512,
+      wedgedRestartMs: 60000,
+      selfRestart: true,
     };
     instance = new LoadShedder({ ...defaultOptions, ...options });
   }

package/src/api/openai-router.js

@@ -426,22 +426,42 @@ async function forwardAnthropicStreamAsOpenAIChunks(res, stream, requestedModel)
             res.write("data: [DONE]\n\n");
             res.end();
             return;
+          case "error":
+            // Forward upstream stream errors in OpenAI wire shape
+            // (data: {"error": ...}) and terminate WITHOUT finish_reason or
+            // [DONE]. Writing "stop" here told agent clients the model chose
+            // to end its turn — they stopped mid-task instead of retrying
+            // (2026-08-19 incident: opencode stalled between tool calls).
+            logger.warn({ error: ev.error?.message }, "Forwarding upstream stream error to OpenAI-surface client");
+            res.write(`data: ${JSON.stringify({ error: { message: ev.error?.message || "upstream stream error", type: ev.error?.type || "server_error", code: null } })}\n\n`);
+            res.end();
+            return;
           default:
             break; // ping, thinking deltas, unknown events — skip
         }
         if (res.writableEnded || res.destroyed) return;
       }
     }
-    // Upstream closed without message_stop — still terminate the client stream cleanly.
+    // Upstream closed without message_stop. If a stop_reason already arrived
+    // the turn is materially complete — finish normally. Otherwise this is a
+    // truncated stream: send an error chunk, never a fabricated "stop".
     if (!res.writableEnded) {
-      res.write(chunk({}, finishReason || "stop"));
-      res.write("data: [DONE]\n\n");
+      if (finishReason) {
+        res.write(chunk({}, finishReason));
+        res.write("data: [DONE]\n\n");
+      } else {
+        logger.warn("Upstream stream ended without message_stop — sending error chunk to OpenAI-surface client");
+        res.write(`data: ${JSON.stringify({ error: { message: "Lynkr: upstream stream truncated — retry", type: "server_error", code: null } })}\n\n`);
+      }
       res.end();
     }
   } catch (err) {
     logger.warn({ err: err.message }, "OpenAI-surface stream forward failed mid-flight");
     if (!res.writableEnded) {
-      try { res.write(chunk({}, "stop")); res.write("data: [DONE]\n\n"); res.end(); } catch { /* client gone */ }
+      try {
+        res.write(`data: ${JSON.stringify({ error: { message: `Lynkr: stream forward failed — retry (${err.message})`, type: "server_error", code: null } })}\n\n`);
+        res.end();
+      } catch { /* client gone */ }
     }
   }
 }
@@ -588,6 +608,34 @@ router.post("/chat/completions", async (req, res) => {
         const content = lynkrBadge(result.body) + (openaiResponse.choices[0].message.content || "");
         let toolCalls = openaiResponse.choices[0].message.tool_calls;
 
+        // Guard: never serve a clean empty completion. Upstream failures
+        // (e.g. exhausted 429 retries) can surface here as a contentless
+        // message with finish_reason "stop"; agent clients read that as
+        // "task complete" and terminate mid-task. Killing the stream without
+        // a finish chunk makes the client retry instead.
+        if (!content && (!toolCalls || toolCalls.length === 0)) {
+          const upstreamErr = result.body?.error || {};
+          logger.error({
+            finishReason: openaiResponse.choices[0]?.finish_reason,
+            terminationReason: result?.terminationReason,
+            status: result?.status,
+            upstreamError: upstreamErr.message || null,
+          }, "Empty completion reached serving path — sending error chunk to client");
+          // res.destroy() here surfaced as ECONNRESET ("connection reset by
+          // server") in clients, and the error chunk raced the teardown so
+          // it often never arrived. A cleanly-ended SSE error chunk (no
+          // finish chunk, no [DONE]) delivers the real upstream error and
+          // still reads as a failed request that agent clients retry.
+          res.write(`data: ${JSON.stringify({ error: {
+            message: upstreamErr.message || "Upstream returned an empty completion; retry.",
+            type: upstreamErr.type || "server_error",
+            code: upstreamErr.code || "empty_completion",
+            ...(result?.status ? { status: result.status } : {}),
+          } })}\n\n`);
+          res.end();
+          return;
+        }
+
         if (clientType !== "unknown" && toolCalls && toolCalls.length > 0) {
           toolCalls = toolCalls.map(tc => {
             const mapped = mapToolForClient(tc.function?.name || "", tc.function?.arguments || "{}", clientType);
@@ -764,6 +812,21 @@ router.post("/chat/completions", async (req, res) => {
         }, "Tool names mapped for non-streaming chat/completions");
       }
 
+      // Guard (mirrors the streaming path): never serve a clean empty
+      // completion — upstream failures can surface as contentless messages
+      // with finish_reason "stop", which agent clients read as "done" and
+      // terminate mid-task. A 502 is retryable; an empty 200 is a silent kill.
+      const _msg = openaiResponse.choices?.[0]?.message;
+      if (!_msg?.content && !(_msg?.tool_calls?.length > 0)) {
+        logger.error({
+          finishReason: openaiResponse.choices?.[0]?.finish_reason,
+          usage: openaiResponse.usage,
+        }, "Empty completion reached non-streaming serving path — returning 502");
+        return res.status(502).json({
+          error: { message: "Upstream returned an empty completion; retry.", type: "server_error", code: "empty_completion" }
+        });
+      }
+
       logger.info({
         duration: Date.now() - startTime,
         mode: "non-streaming",
@@ -939,6 +1002,18 @@ function getConfiguredProviders() {
     });
   }
 
+  if (config.baidu?.apiKey) {
+    providers.push({
+      name: "baidu",
+      type: "baidu-qianfan",
+      models: [
+        config.baidu.model || "ernie-4.5-turbo-128k",
+        "ernie-4.5-turbo-128k",
+        "ernie-x1.1"
+      ]
+    });
+  }
+
   if (config.vertex?.projectId) {
     providers.push({
       name: "vertex",

package/src/api/providers-handler.js

@@ -107,6 +107,19 @@ function getConfiguredProviders() {
     });
   }
 
+  // Check Atlas Cloud (OpenAI-compatible Chat Completions)
+  if (config.atlas?.apiKey) {
+    providers.push({
+      name: "atlas",
+      type: "atlas-cloud",
+      baseUrl: config.atlas.endpoint || "https://api.atlascloud.ai/v1/chat/completions",
+      enabled: true,
+      models: [
+        { id: config.atlas.model || "qwen/qwen3.8-max", name: "Configured Model" },
+      ]
+    });
+  }
+
   // Check OpenRouter
   if (config.openrouter?.apiKey) {
     providers.push({
@@ -210,6 +223,21 @@ function getConfiguredProviders() {
     });
   }
 
+  // Check Baidu Qianfan (ERNIE)
+  if (config.baidu?.apiKey) {
+    providers.push({
+      name: "baidu",
+      type: "baidu-qianfan",
+      baseUrl: config.baidu.endpoint || "https://qianfan.baidubce.com/v2/chat/completions",
+      enabled: true,
+      models: [
+        { id: config.baidu.model || "ernie-4.5-turbo-128k", name: "Configured Model" },
+        { id: "ernie-4.5-turbo-128k", name: "ERNIE 4.5 Turbo 128K" },
+        { id: "ernie-x1.1", name: "ERNIE X1.1 (reasoning)" },
+      ]
+    });
+  }
+
   // Check Vertex AI (Google Cloud)
   if (config.vertex?.projectId) {
     const region = config.vertex.region || "us-east5";