Schema | model-council-mcp

model-council-mcp

Describes the environment variables required to run the server.

Name	Required	Description	Default
`JUDGE_MODEL`	No	Judge model ID or 'auto' to use largest council member (default: auto)	auto
`GROQ_API_KEY`	No	Groq API key to enable Groq models (optional)
`VLLM_SERVERS`	No	Comma-separated list of vLLM servers (name:host:port or name:url format, optional)
`RESPONSE_MODE`	No	Default response mode: individual, categorized, or deconflicted (default: categorized)	categorized
`COUNCIL_MODELS`	No	Comma-separated list of model IDs for the council (empty means auto-discover)
`OLLAMA_ADDRESS`	No	Base URL of your Ollama server (default: http://localhost:11434)	http://localhost:11434
`OPENAI_API_KEY`	No	OpenAI API key to enable OpenAI models (optional)
`SGLANG_SERVERS`	No	Comma-separated list of SGLang servers (name:host:port or name:url format, optional)
`TRTLLM_SERVERS`	No	Comma-separated list of TRT-LLM servers (name:host:port or name:url format, optional)
`ANTHROPIC_API_KEY`	No	Anthropic API key to enable Claude models (optional)
`MAX_DECONFLICT_ROUNDS`	No	Maximum deconfliction iterations (1-10, default: 3)	3

Features and capabilities supported by this server

Capability	Details
`tools`	{}

Functions exposed to the LLM to take actions

Name	Description
list_modelsA	List all AI models available across every configured provider (Ollama, OpenAI, Anthropic, Groq, vLLM, TRT-LLM, SGLang). Use the returned model IDs when calling configure_council.
configure_councilB	Update the council configuration: select which models form the council, choose a judge model, set the response mode (individual / categorized / deconflicted), and set the maximum deconfliction rounds.
ask_councilA	Send a question to the model council and get a structured response. Mode: individual (each model answers separately), categorized (judge groups responses into agreement/complementary/conflicting), or deconflicted (iterative loop — judge orchestrates re-questioning until conflicts resolve, returns a deconfliction score 0–100%).
get_council_configA	Return the current council configuration: member models, judge model, response mode, and max deconfliction rounds.

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

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

curl -X GET 'https://glama.ai/api/mcp/v1/servers/tsarihan/model-council-mcp'

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