Skip to main content
Glama
BlockRunAI

BlockRun MCP

Official
by BlockRunAI
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Hybrid

blockrun_chat

Get a second opinion from another AI model or route a task to a specialized model. Choose coding, reasoning, budget, or free modes, or specify a custom model for the job.

Instructions

Get a second opinion from another AI model, or use a specialized model for a specific task.

Notable modes:

  • mode:"powerful" → Claude Opus 5, Claude Opus 4.8, GPT-5.6-sol, Claude Fable 5 (frontier, 1M context)

  • mode:"reasoning" → Claude Opus 5, GPT-5.6-sol, Kimi K3, Grok 4.3, deepseek-v4-pro

  • mode:"coding" → Claude Opus 5, GPT-5.3-codex, Kimi K3, Grok Build, GLM-5.2

  • mode:"cheap" → deepseek-v4-pro, Qwen3.7 Flash, MiniMax M3, Tencent Hy3

  • mode:"glm" → Zhipu GLM-5 / 5.2 / 5.1 / 5-Turbo (strong at coding)

  • mode:"free" → NVIDIA models (no cost)

Pick directly: model:"anthropic/claude-opus-5", model:"moonshot/kimi-k3", model:"openai/gpt-5.6-sol", model:"xai/grok-4.5", model:"nvidia/gpt-oss-120b" (free).

Run blockrun_models to see all available models with pricing.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
modeNoRouting mode: powerful/reasoning = frontier models (Opus 5, GPT-5.6-sol, Kimi K3), coding = code-specialized, glm = Zhipu GLM (great for coding), cheap = budget models, free = NVIDIA only (ignored if model specified)
stopNoUp to 4 stop sequences; generation halts when any is produced
modelNoSpecific model ID (e.g., 'moonshot/kimi-k3', 'openai/gpt-5.6-sol', 'zai/glm-5')
systemNoOptional system prompt
messageYesYour message to the AI
agent_idNoAgent 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.
messagesNoConversation history for multi-turn context. When provided, 'message' is appended as the final user turn. Use with explicit 'model' param (defaults to 'openai/gpt-5.6-terra' 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.
thinkingNoAnthropic extended thinking. Only honored for anthropic/claude-* models — these go direct to the native /v1/messages endpoint and the response includes verbatim type:'thinking' blocks with their original signature. Ignored for non-Claude models (no native thinking channel).
max_tokensNoMax tokens in response
temperatureNoCreativity 0-2
response_formatNoSet to 'json_object' to force valid JSON output (no markdown fences). Works across all providers.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. Changed1 schema field changedv0.45.1
    • removedInput schema / $schema
      Removed value: -"http://json-schema.org/draft-07/schema#"
  2. Changed1 schema field changedv0.40.1
    • changedInput schema / properties / mode / description
      Previous value: -"Routing mode: powerful/reasoning = frontier models (Opus 4.8, GPT-5.6-sol, Kimi K3), coding = code-specialized, glm = Zhipu GLM (cheap, great for coding), cheap = budget models, free = NVIDIA only (ignored if model specified)"New value: +"Routing mode: powerful/reasoning = frontier models (Opus 5, GPT-5.6-sol, Kimi K3), coding = code-specialized, glm = Zhipu GLM (great for coding), cheap = budget models, free = NVIDIA only (ignored if model specified)"
  3. Changed3 schema fields changedv0.33.0
    • changedInput schema / properties / messages / description
      Previous value: -"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-5.5' 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."New value: +"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-5.6-terra' 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."
    • changedInput schema / properties / mode / description
      Previous value: -"Routing mode: glm = Zhipu GLM-5/GLM-5-Turbo ($0.001/call, great for coding), coding = GLM-5 + code models, cheap = GLM-5 + budget, free = NVIDIA only (ignored if model specified)"New value: +"Routing mode: powerful/reasoning = frontier models (Opus 4.8, GPT-5.6-sol, Kimi K3), coding = code-specialized, glm = Zhipu GLM (cheap, great for coding), cheap = budget models, free = NVIDIA only (ignored if model specified)"
    • changedInput schema / properties / model / description
      Previous value: -"Specific model ID (e.g., 'zai/glm-5', 'openai/o3')"New value: +"Specific model ID (e.g., 'moonshot/kimi-k3', 'openai/gpt-5.6-sol', 'zai/glm-5')"
  4. Changed2 schema fields changedv0.31.6
    • removedInput schema / properties / routing
      Removed value: -{
-  "description": "Set to \"smart\" to auto-select the optimal model via ClawRouter (14-dimension AI routing)",
-  "enum": [
-    "smart"
-  ],
-  "type": "string"
-}
    • removedInput schema / properties / routing_profile
      Removed value: -{
-  "default": "auto",
-  "description": "Cost/quality profile for ClawRouter: \"eco\" (budget), \"auto\" (balanced, default), \"premium\" (best quality). Note: \"free\" maps to \"auto\" (the SDK dropped the free profile) and still settles a PAID model — for zero-cost generation use mode:\"free\" or model:\"nvidia/...\". Only applies when routing:\"smart\".",
-  "enum": [
-    "free",
-    "eco",
-    "auto",
-    "premium"
-  ],
-  "type": "string"
-}
  5. Changed5 schema fields changedv0.25.2
    • changedInput schema / properties / routing_profile / description
      Previous value: -"Cost/quality profile for ClawRouter: \"free\" (zero cost NVIDIA), \"eco\" (budget), \"auto\" (balanced, default), \"premium\" (best quality) (only applies when routing: \"smart\")"New value: +"Cost/quality profile for ClawRouter: \"eco\" (budget), \"auto\" (balanced, default), \"premium\" (best quality). Note: \"free\" maps to \"auto\" (the SDK dropped the free profile) and still settles a PAID model — for zero-cost generation use mode:\"free\" or model:\"nvidia/...\". Only applies when routing:\"smart\"."
    • changedInput schema / properties / thinking / properties / budget_tokens / description
      Previous value: -"Tokens Claude may spend reasoning before answering. max_tokens is auto-raised above this if needed."New value: +"Tokens Claude may spend reasoning before answering (1024–100000; Anthropic requires ≥1024). max_tokens is auto-raised above this if needed; counts toward the budget reserve."
    • addedInput schema / properties / thinking / properties / budget_tokens / maximum
      Added value: +100000
    • changedInput schema / properties / thinking / properties / budget_tokens / minimum
      Previous value: -1New value: +1024
    • changedInput schema / properties / thinking / properties / budget_tokens / type
      Previous value: -"number"New value: +"integer"
  6. Changed7 schema fields changedv0.22.0
    • changedInput schema / properties / messages / description
      Previous value: -"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."New value: +"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-5.5' 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."
    • addedInput schema / properties / messages / items / properties / content / anyOf
      Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "items": {
+      "anyOf": [
+        {
+          "properties": {
+            "text": {
+              "type": "string"
+            },
+            "type": {
+              "const": "text",
+              "type": "string"
+            }
+          },
+          "required": [
+            "type",
+            "text"
+          ],
+          "type": "object"
+        },
+        {
+          "properties": {
+            "image_url": {
+              "properties": {
+                "url": {
+                  "description": "https URL or data:<mime>;base64,<...> URI",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "type": "object"
+            },
+            "type": {
+              "const": "image_url",
+              "type": "string"
+            }
+          },
+          "required": [
+            "type",
+            "image_url"
+          ],
+          "type": "object"
+        }
+      ]
+    },
+    "type": "array"
+  }
+]
    • addedInput schema / properties / messages / items / properties / content / description
      Added value: +"Plain text, or an array of parts for multimodal input (text + image_url). Images are honored on the native anthropic/claude-* path."
    • removedInput schema / properties / messages / items / properties / content / type
      Removed value: -"string"
    • addedInput schema / properties / response_format
      Added value: +{
+  "description": "Set to 'json_object' to force valid JSON output (no markdown fences). Works across all providers.",
+  "enum": [
+    "text",
+    "json_object"
+  ],
+  "type": "string"
+}
    • addedInput schema / properties / stop
      Added value: +{
+  "description": "Up to 4 stop sequences; generation halts when any is produced",
+  "items": {
+    "type": "string"
+  },
+  "maxItems": 4,
+  "type": "array"
+}
    • addedInput schema / properties / thinking
      Added value: +{
+  "description": "Anthropic extended thinking. Only honored for anthropic/claude-* models — these go direct to the native /v1/messages endpoint and the response includes verbatim type:'thinking' blocks with their original signature. Ignored for non-Claude models (no native thinking channel).",
+  "properties": {
+    "budget_tokens": {
+      "description": "Tokens Claude may spend reasoning before answering. max_tokens is auto-raised above this if needed.",
+      "minimum": 1,
+      "type": "number"
+    },
+    "type": {
+      "const": "enabled",
+      "type": "string"
+    }
+  },
+  "required": [
+    "type",
+    "budget_tokens"
+  ],
+  "type": "object"
+}
  7. First observedv0.16.2

TDQS

A4.1/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations are minimal (readOnlyHint=false, destructiveHint=false), so the description carries some burden. It adds meaningful behavioral context: mode-to-model mappings, the '1M context' claim for frontier models, and the cost implication via 'free' (no cost) plus the pointer to pricing. It does not disclose output format, rate limits, or that this is a live paid API call, but it does not contradict the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The purpose is front-loaded in the first sentence, followed by a scannable bulleted mode list, then direct model picks, then a single pointer to the sibling for pricing. Every sentence earns its place; there is no filler or repetition. The length is justified by the routing information it conveys.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

This is a complex tool (11 parameters, nested messages/thinking objects) with no output schema, so the description should explain return values and edge conditions. It covers model routing well but does not describe what the response looks like (e.g., text vs. thinking blocks) or explicitly route non-text tasks to media siblings. The schema's 100% parameter coverage compensates for some gaps, but the lack of output guidance leaves the definition incomplete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3, but the description adds significant value by mapping the 'mode' parameter to concrete model pools and listing valid 'model' IDs like 'anthropic/claude-opus-5' and 'moonshot/kimi-k3'. These examples and mode→model mappings go beyond the schema's brief descriptions. Other parameters (stop, temperature, thinking) are already well-covered by the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb and resource: 'Get a second opinion from another AI model, or use a specialized model for a specific task.' This clearly identifies the tool as the LLM text-generation/chat member of the blockrun family, distinguishing it from the image, music, video, wallet, and search siblings. The mode list further pins down the scope without ambiguity.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives explicit routing guidance by mapping modes to use cases (powerful/reasoning for frontier models, coding for code-specialized, cheap for budget, free for no-cost) and tells the agent to 'Run blockrun_models to see all available models with pricing,' naming the sibling for model discovery. It does not explicitly state when not to use this tool versus media-generation siblings, but the mode guidance and context make the intended usage clear.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/BlockRunAI/blockrun-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server