What this kind of tool is for
- Hand your agent a structured way to call a model or an inference API.
- Compare outputs from two different providers on the same prompt.
- Wire retrieval, embeddings, or generation into an existing agent workflow.
How to compare servers like this
- Provider coverage:
- Match the tool's supported providers and models to the ones your project uses.
- Cost exposure:
- Model-backed tools spend real API budget; check how usage and limits are configured.
- Maintenance signal:
- The provider landscape shifts weekly; prefer tools updated within the last few months.
- Config surface:
- Prefer tools whose environment-variable or config keys are documented plainly.
This page carries your copy-paste config: pick your MCP client in the install section on this page and use the copy button.
What is the io.github.faulkj/legion MCP server?
MCP-native LLM councils for debates, juries, blind panels, refinement, and custom deliberation.
How to install io.github.faulkj/legion
Copy-paste configuration for popular MCP clients.
MCP_TRANSPORTrequiredRun the installed package over MCP stdio (the binary defaults to HTTP).
DEFAULT_BASE_URLDefault OpenAI Responses-compatible API root for model files that omit baseUrl.
DEFAULT_API_KEYsecretDefault provider or gateway key for model files that omit apiKey.
ALLOW_NO_MODELSrequiredBoot without any config/models/*.json. The package ships only key-free examples, which the scanner ignores, so this must stay true until a real model file exists beside the working directory.
OPENAI_API_KEYsecretOptional. A model file's apiKey may be "env:VAR" to read the key from an environment variable named VAR at startup; name those variables here as needed. The bundled gpt example reads this one.
MAX_ROUNDSMaximum discussion rounds accepted by a council run.
TOKEN_BUDGETOptional soft cumulative token budget for a council run.
DYNAMIC_ROLESAllow callers to define ad-hoc council roles inline.
LOG_LEVELLogging verbosity. Logs are written to stderr for stdio safety.
{
"mcpServers": {
"legion": {
"command": "npx",
"args": [
"-y",
"legion-mcp"
],
"env": {
"MCP_TRANSPORT": "<YOUR_MCP_TRANSPORT>",
"DEFAULT_BASE_URL": "<YOUR_DEFAULT_BASE_URL>",
"DEFAULT_API_KEY": "<YOUR_DEFAULT_API_KEY>",
"ALLOW_NO_MODELS": "<YOUR_ALLOW_NO_MODELS>",
"OPENAI_API_KEY": "<YOUR_OPENAI_API_KEY>",
"MAX_ROUNDS": "<YOUR_MAX_ROUNDS>",
"TOKEN_BUDGET": "<YOUR_TOKEN_BUDGET>",
"DYNAMIC_ROLES": "<YOUR_DYNAMIC_ROLES>",
"LOG_LEVEL": "<YOUR_LOG_LEVEL>"
}
}
}
}Related MCP servers

FHIR MCP server with SMART Backend Services, FHIRPath filtering, and configurable clinical data access.
Kojiru Recourse Layer — agent credit primitives (ACS, stake, draw, anchor) live on Base mainnet
View repository →
OpenTranscribe MCP
Provider-independent speech-to-text with normalized transcripts over MCP.

FeedOracle Stablecoin Risk
7-signal stablecoin risk scoring. 13 tools, 28+ tokens, SAFE/CAUTION/AVOID verdicts.

Let AI agents inspect, debug, and drive running WPF apps—visual tree, bindings, screenshots, clicks, and keyboard input.

MaxStat MCP
Search 408K+ MAX channels and 86M+ posts; metrics, reactions, forwards, webhooks and usage.
