@blockrun/mcp 0.16.1 → 0.17.0

package/dist/index.js

@@ -214,11 +214,11 @@ var BASE_CHAIN_ID = "8453";
 var MODEL_TIERS = {
   fast: ["google/gemini-3.5-flash", "google/gemini-2.5-flash", "google/gemini-3.1-flash-lite", "openai/gpt-5-mini", "deepseek/deepseek-chat", "google/gemini-3-flash-preview"],
   balanced: ["openai/gpt-5.4", "anthropic/claude-sonnet-4.6", "google/gemini-2.5-pro", "moonshot/kimi-k2.6", "openai/gpt-5.3", "google/gemini-3.1-pro"],
-  powerful: ["openai/gpt-5.4-pro", "anthropic/claude-opus-4.7", "anthropic/claude-opus-4.6", "openai/o3", "openai/gpt-5.4"],
+  powerful: ["anthropic/claude-opus-4.8", "openai/gpt-5.4-pro", "anthropic/claude-opus-4.7", "anthropic/claude-opus-4.6", "openai/o3", "openai/gpt-5.4"],
   cheap: ["zai/glm-5", "zai/glm-5-turbo", "nvidia/gpt-oss-120b", "nvidia/deepseek-v3.2", "google/gemini-2.5-flash", "deepseek/deepseek-chat", "openai/gpt-5.4-nano"],
-  reasoning: ["openai/o3", "openai/o1", "openai/o3-mini", "deepseek/deepseek-reasoner", "moonshot/kimi-k2.6", "openai/gpt-5.3-codex"],
+  reasoning: ["anthropic/claude-opus-4.8", "openai/o3", "openai/o1", "openai/o3-mini", "deepseek/deepseek-reasoner", "moonshot/kimi-k2.6", "openai/gpt-5.3-codex"],
   free: ["nvidia/qwen3-next-80b-a3b-thinking", "nvidia/mistral-small-4-119b", "nvidia/gpt-oss-120b", "nvidia/deepseek-v3.2", "nvidia/qwen3-coder-480b", "nvidia/llama-4-maverick", "nvidia/gpt-oss-20b", "nvidia/glm-4.7"],
-  coding: ["zai/glm-5", "openai/gpt-5.3-codex", "moonshot/kimi-k2.6", "nvidia/qwen3-coder-480b", "anthropic/claude-sonnet-4.6", "openai/gpt-5.4"],
+  coding: ["anthropic/claude-opus-4.8", "zai/glm-5", "openai/gpt-5.3-codex", "moonshot/kimi-k2.6", "nvidia/qwen3-coder-480b", "anthropic/claude-sonnet-4.6", "openai/gpt-5.4"],
   glm: ["zai/glm-5", "zai/glm-5-turbo", "nvidia/glm-4.7"]
 };
 
@@ -714,6 +714,8 @@ Run blockrun_models to see all 41+ models with pricing.`,
         system: z2.string().optional().describe("Optional system prompt"),
         max_tokens: z2.number().optional().default(1024).describe("Max tokens in response"),
         temperature: z2.number().optional().default(1).describe("Creativity 0-2"),
+        response_format: z2.enum(["text", "json_object"]).optional().describe("Set to 'json_object' to force valid JSON output (no markdown fences). Works across all providers."),
+        stop: z2.array(z2.string()).max(4).optional().describe("Up to 4 stop sequences; generation halts when any is produced"),
         agent_id: z2.string().optional().describe("Agent identifier. If a budget was delegated for this agent_id via blockrun_wallet action:'delegate', spending is tracked and enforced. The agent is hard-stopped when its budget is exhausted."),
         messages: z2.array(z2.object({
           role: z2.enum(["user", "assistant", "system"]),
@@ -721,8 +723,9 @@ Run blockrun_models to see all 41+ models with pricing.`,
         })).optional().describe("Conversation history for multi-turn context. When provided, 'message' is appended as the final user turn. Use with explicit 'model' param (defaults to 'openai/gpt-5.4' if not specified). Note: if you include a role:'system' entry in messages[], do not also pass the system param to avoid duplicate system messages.")
       }
     },
-    async ({ message, model, mode, routing, routing_profile, system, max_tokens, temperature, agent_id, messages }) => {
+    async ({ message, model, mode, routing, routing_profile, system, max_tokens, temperature, response_format, stop, agent_id, messages }) => {
       const llm = getClient();
+      const responseFormat = response_format ? { type: response_format } : void 0;
       const estimatedCost = estimateChatCost(mode, model, routing, routing_profile);
       const budgetCheck = checkBudget(budget, agent_id, estimatedCost);
       if (!budgetCheck.allowed) {
@@ -744,7 +747,12 @@ Run blockrun_models to see all 41+ models with pricing.`,
             maxTokens: max_tokens,
             maxOutputTokens: max_tokens,
             temperature,
-            routingProfile: routing_profile
+            // @blockrun/llm 2.x dropped the "free" routing profile; the gateway
+            // already routes to the most cost-effective model by default, so we
+            // omit it and let ClawRouter pick (matches the SDK upgrade path).
+            routingProfile: routing_profile === "free" ? void 0 : routing_profile,
+            responseFormat,
+            stop
           });
           recordSpending(budget, result.routing.costEstimate || 1e-3, agent_id);
           return {
@@ -772,7 +780,9 @@ ${result.response}` }],
         try {
           const result = await llm.chatCompletion(targetModel, fullMessages, {
             maxTokens: max_tokens,
-            temperature
+            temperature,
+            responseFormat,
+            stop
           });
           const reply = result.choices?.[0]?.message?.content || "";
           recordSpending(budget, estimatedCost, agent_id);
@@ -792,7 +802,9 @@ ${reply}` }],
           const response = await llm.chat(model, message, {
             system,
             maxTokens: max_tokens,
-            temperature
+            temperature,
+            responseFormat,
+            stop
           });
           recordSpending(budget, estimatedCost, agent_id);
           return { content: [{ type: "text", text: response }] };
@@ -812,7 +824,9 @@ ${reply}` }],
           const response = await llm.chat(m, message, {
             system,
             maxTokens: max_tokens,
-            temperature
+            temperature,
+            responseFormat,
+            stop
           });
           recordSpending(budget, estimatedCost, agent_id);
           return {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@blockrun/mcp",
-  "version": "0.16.1",
+  "version": "0.17.0",
   "mcpName": "io.github.BlockRunAI/blockrun-mcp",
   "description": "BlockRun MCP Server - Give your AI agent web search, deep research, prediction markets, crypto data, X/Twitter intelligence. Paid via x402 micropayments.",
   "type": "module",
@@ -44,7 +44,7 @@
     "url": "https://github.com/blockrunai/blockrun-mcp/issues"
   },
   "dependencies": {
-    "@blockrun/llm": "^1.12.0",
+    "@blockrun/llm": "^2.11.0",
     "@modelcontextprotocol/sdk": "^1.0.0",
     "open": "^11.0.0",
     "qrcode": "^1.5.4",