Skip to main content
Glama
Platano78

Smart-AI-Bridge

by Platano78

ask

Send one prompt to any AI backend and get the response. Auto routing picks the best model by complexity, or force a specific provider.

Instructions

Send one prompt to one AI backend and return the response. model:'auto' lets SAB's router pick the best backend by task complexity + current health; passing a specific model name forces that provider. Use this for direct LLM queries that don't fit a more specialized tool. For multi-backend consensus on the same prompt, use council. For agentic multi-step work with a defined role, use spawn_subagent. For LLM-driven file generation or editing, use generate_file / modify_file so the file content stays out of Claude's context window. Read-only: makes one HTTP call to the chosen backend. Returns: {success, model, requested_backend, actual_backend, prompt (truncated preview), response (the LLM output), backend_used, fallback_chain, response_time, cache_status, thinking_enabled, max_tokens, was_truncated, smart_routing_applied, routing, processing_time}.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
modelYesAI backend to query: auto (smart routing selects a lane by task complexity + current health), local (your own router — vLLM/llama.cpp/LM Studio — autodiscovered), gemini (Google Gemini lane), nvidia_deepseek (NVIDIA-hosted DeepSeek lane — reasoning-oriented, supports streaming and the `thinking` option), nvidia_glm (NVIDIA-hosted GLM lane — code-oriented), openai (OpenAI lane), groq (Groq lane — low-latency hosted inference). No model id or context size is fixed here: each lane runs whatever `config.model` declares in backends.json, or a model selected from the provider's own catalog when nothing is declared. The friendly aliases `deepseek`, `glm` and `openai` are also accepted (mapped to nvidia_deepseek / nvidia_glm / openai_chatgpt), matching the other tools.
promptYesYour question or prompt (Unity/complex generations automatically get high token limits)
thinkingNoEnable thinking mode for DeepSeek (shows reasoning)
max_tokensNoMaximum response length (auto-calculated if not specified: Unity=16K, Complex=8K, Simple=2K)
force_backendNoForce specific backend (bypasses smart routing) - use backend keys like "local", "gemini", "nvidia_deepseek", "nvidia_glm", "openai_chatgpt", "groq"
model_profileNoOnly for the local backend: request a specific model id from your local router (llama-swap/llama.cpp/LM Studio). Valid values are whatever your own router serves — query its /v1/models endpoint for the list; SAB does not ship a fixed set. The id is passed through as the requested model, so an id your router does not serve fails the call (HTTP 400 model not found) rather than falling back. SAB also tries to load the model first if it is not resident, but that pre-flight step only probes http://localhost:8081 — if your server listens on another port the load is skipped (the request still goes to your configured endpoint). Omit to use whichever model is already loaded.
enable_chunkingNoEnable automatic request chunking for extremely large generations (fallback if truncated)

Schema Changelog

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

  1. Changed2 schema fields changedv2.15.0
    • changedInput schema / properties / force_backend / description
      Previous value: -"Force specific backend (bypasses smart routing) - use backend keys like \"local\", \"gemini\", \"nvidia_deepseek\", \"nvidia_glm\", \"openai\", \"groq\""New value: +"Force specific backend (bypasses smart routing) - use backend keys like \"local\", \"gemini\", \"nvidia_deepseek\", \"nvidia_glm\", \"openai_chatgpt\", \"groq\""
    • changedInput schema / properties / model / description
      Previous value: -"AI backend to query: auto (smart routing selects optimal backend), local (autodiscover vLLM/llama.cpp/LM Studio), gemini (Gemini Enhanced, 32K tokens), nvidia_deepseek (NVIDIA DeepSeek with streaming + reasoning, 8K tokens), nvidia_glm (NVIDIA GLM-5.2 code specialist, 32K tokens), openai (OpenAI GPT-5.2, 128K context, premium reasoning), groq (Llama 3.3 70B, ultra-fast 500+ t/s). The friendly aliases `deepseek` and `glm` are also accepted (mapped to nvidia_deepseek / nvidia_glm), matching the other tools."New value: +"AI backend to query: auto (smart routing selects a lane by task complexity + current health), local (your own router — vLLM/llama.cpp/LM Studio — autodiscovered), gemini (Google Gemini lane), nvidia_deepseek (NVIDIA-hosted DeepSeek lane — reasoning-oriented, supports streaming and the `thinking` option), nvidia_glm (NVIDIA-hosted GLM lane — code-oriented), openai (OpenAI lane), groq (Groq lane — low-latency hosted inference). No model id or context size is fixed here: each lane runs whatever `config.model` declares in backends.json, or a model selected from the provider's own catalog when nothing is declared. The friendly aliases `deepseek`, `glm` and `openai` are also accepted (mapped to nvidia_deepseek / nvidia_glm / openai_chatgpt), matching the other tools."
  2. Changed3 schema fields changedv2.14.0
    • changedInput schema / properties / force_backend / description
      Previous value: -"Force specific backend (bypasses smart routing) - use backend keys like \"local\", \"gemini\", \"nvidia_deepseek\", \"nvidia_glm\", \"openai\", \"groq\" (legacy \"nvidia_qwen\"/\"qwen3\" also accepted)"New value: +"Force specific backend (bypasses smart routing) - use backend keys like \"local\", \"gemini\", \"nvidia_deepseek\", \"nvidia_glm\", \"openai\", \"groq\""
    • changedInput schema / properties / model / description
      Previous value: -"AI backend to query: auto (smart routing selects optimal backend), local (autodiscover vLLM/llama.cpp/LM Studio), gemini (Gemini Enhanced, 32K tokens), nvidia_deepseek (NVIDIA DeepSeek with streaming + reasoning, 8K tokens), nvidia_glm (NVIDIA GLM-5.2 code specialist, 32K tokens), openai (OpenAI GPT-5.2, 128K context, premium reasoning), groq (Llama 3.3 70B, ultra-fast 500+ t/s). The friendly aliases `deepseek` and `glm` are also accepted (mapped to nvidia_deepseek / nvidia_glm), matching the other tools. `nvidia_qwen` and `qwen3` are legacy aliases still accepted for back-compat (the lane served Qwen3 Coder 480B until NVIDIA retired it on 2026-06-11) — they resolve to nvidia_glm."New value: +"AI backend to query: auto (smart routing selects optimal backend), local (autodiscover vLLM/llama.cpp/LM Studio), gemini (Gemini Enhanced, 32K tokens), nvidia_deepseek (NVIDIA DeepSeek with streaming + reasoning, 8K tokens), nvidia_glm (NVIDIA GLM-5.2 code specialist, 32K tokens), openai (OpenAI GPT-5.2, 128K context, premium reasoning), groq (Llama 3.3 70B, ultra-fast 500+ t/s). The friendly aliases `deepseek` and `glm` are also accepted (mapped to nvidia_deepseek / nvidia_glm), matching the other tools."
    • changedInput schema / properties / model / enum
      Previous value: -[
      -  "auto",
      -  "local",
      -  "gemini",
      -  "groq",
      -  "deepseek",
      -  "glm",
      -  "qwen3",
      -  "nvidia_deepseek",
      -  "nvidia_glm",
      -  "nvidia_qwen",
      -  "openai"
      -]New value: +[
      +  "auto",
      +  "local",
      +  "gemini",
      +  "groq",
      +  "deepseek",
      +  "glm",
      +  "nvidia_deepseek",
      +  "nvidia_glm",
      +  "openai"
      +]
  3. Added
  4. Removedv2.8.1
  5. Changed2 schema fields changedv2.6.0
    • changedInput schema / properties / model / description
      Previous value: -"AI backend to query: auto (smart routing selects optimal backend), local (autodiscover vLLM/llama.cpp/LM Studio), gemini (Gemini Enhanced, 32K tokens), nvidia_deepseek (NVIDIA DeepSeek with streaming + reasoning, 8K tokens), nvidia_qwen (NVIDIA Qwen3 Coder 480B, 32K tokens), openai (OpenAI GPT-5.2, 128K context, premium reasoning), groq (Llama 3.3 70B, ultra-fast 500+ t/s)"New value: +"AI backend to query: auto (smart routing selects optimal backend), local (autodiscover vLLM/llama.cpp/LM Studio), gemini (Gemini Enhanced, 32K tokens), nvidia_deepseek (NVIDIA DeepSeek with streaming + reasoning, 8K tokens), nvidia_qwen (NVIDIA Qwen3 Coder 480B, 32K tokens), openai (OpenAI GPT-5.2, 128K context, premium reasoning), groq (Llama 3.3 70B, ultra-fast 500+ t/s). The friendly aliases `deepseek` and `qwen3` are also accepted (mapped to nvidia_deepseek / nvidia_qwen), matching the other tools."
    • changedInput schema / properties / model / enum
      Previous value: -[
      -  "auto",
      -  "local",
      -  "gemini",
      -  "nvidia_deepseek",
      -  "nvidia_qwen",
      -  "openai",
      -  "groq"
      -]New value: +[
      +  "auto",
      +  "local",
      +  "gemini",
      +  "groq",
      +  "deepseek",
      +  "qwen3",
      +  "nvidia_deepseek",
      +  "nvidia_qwen",
      +  "openai"
      +]
  6. Addedv1.3.2
  7. Removedv1.0.0
  8. First observed

TDQS

A4.6/5.0
Behavior4/5

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

With no annotations provided, the description carries the full burden of behavioral disclosure. It states 'Read-only: makes one HTTP call to the chosen backend' and details the return object fields, including routing, fallback_chain, and cache_status. This is strong coverage; however, it does not mention rate limits, authentication requirements, or error handling beyond the model_profile note, so a full 5 is not warranted.

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

Conciseness4/5

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

The description is long but every sentence carries distinct value—core purpose, routing explanation, alternative routing, read-only guarantee, and return structure. It is front-loaded with the main purpose and scoping. Slightly overlong relative to typical descriptions, but no redundancy. A 4 rather than 5 because it could be trimmed around the model-profile paragraph.

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

Completeness5/5

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

For a tool with 7 parameters, 2 required, and no output schema, the description is exceptionally complete. It covers usage, alternatives, parameter details (especially model routing and model_profile), and explicitly lists the return fields, compensating for the absence of an output schema. An agent has everything needed to invoke this tool correctly.

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%, giving a baseline of 3. The description adds substantial meaning beyond the schema: it clarifies the 'auto' routing behavior, explains model aliases (deepseek, glm, openai) and their mappings, details model_profile's behavior with local routers and the fallback caveat, and describes chunking. This elevates it above baseline.

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 precise verb+resource statement: 'Send one prompt to one AI backend and return the response.' It then explicitly contrasts with sibling tools (council, spawn_subagent, generate_file, modify_file) by naming each alternative and its purpose, making differentiation immediate.

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

Usage Guidelines5/5

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

Usage guidance is explicit and actionable: 'Use this for direct LLM queries that don't fit a more specialized tool.' It then enumerates specific sibling tools and the conditions that route elsewhere (multi-backend consensus -> council; agentic multi-step -> spawn_subagent; file generation/editing -> generate_file/modify_file). No ambiguity remains.

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

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/Platano78/Smart-AI-Bridge'

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