MCP SERVERPut many MCP servers behind one connection: search across all their tools in natural language, and call the one you found — without every tool definition sitting in context.
LABSCO SUMMARY
The point is context, not convenience: connect ten servers directly and their definitions travel with every turn, whereas here the model carries a search call and pulls a definition only when it needs one. That trade has a cost — a step of searching before acting, and results only as good as the descriptions the downstream servers wrote. Redis as the storage backend is what makes the tool index survive a restart rather than being rebuilt.
A hub that registers downstream MCP servers from a config file, indexes their tools, and exposes two calls of its own: one to find a tool, one to invoke it. Tool names are namespaced as server_name__tool_name, so collisions between servers cannot happen.
pip install mcp-orchestrator, then run mcp-orchestrator; nothing to authorize for the hub itself, and credentials for downstream servers live in their entries in server_config.json, whose path SERVER_CONFIG_PATH overrides. STORAGE_BACKEND selects memory or redis, with REDIS_URL pointing at the instance. ORCHESTRATOR_TRANSPORT=http and ORCHESTRATOR_PORT switch it to HTTP, serving at http://localhost:8080/mcp by default. MCP_ORCHESTRATOR_TOOL_CACHE_TTL defaults to 300 seconds, MCP_ORCHESTRATOR_CONNECTION_TIMEOUT to 30.0 and MCP_ORCHESTRATOR_MAX_RETRIES to 3. MIT.
One command — pip install mcp-orchestrator
USE CASES
Register them once in the config; the client sees a search call rather than every tool from every server, and definitions load only when one matches.
Describe the job in plain words and get ranked matches across all registered servers, with names, descriptions and arguments.
Switch the same search to regex when you want exact matches across names and descriptions rather than relevance ranking.
Invoke it by its namespaced server_name__tool_name with your arguments; the downstream server's result comes back untouched.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
call_remote_toolCall a tool directly on a registered remote MCP server through the orchestrator.
This tool allows direct invocation of any tool on a downstream MCP server. The tool name should be in the format 'server_name__tool_name' (e.g., 'context7__query-docs').
tool_searchSearch for tools using BM25 relevance ranking or regex pattern matching.
Searches across tool names, descriptions, and argument names/descriptions. By default, uses BM25 natural language search. Set use_regex=True to search using Python regex patterns instead. Returns tool_reference blocks for discovered tools with deferred loading.
BEFORE YOU INSTALL
2 tools when we connected; the project documents 3. Downstream tools register only after tool_search finds them, so the list grows during a session. discover_tools is documented but absent from version 0.1.2.
The documented client runs the server on your machine.
INSTALLATION
pip install mcp-orchestrator{
"mcpServers": {
"mcp-orchestrator": {
"command": "mcp-orchestrator",
"env": {
"STORAGE_BACKEND": "memory",
"ORCHESTRATOR_LOG_LEVEL": "INFO"
}
}
}
}MCP servers
gybob
1mcp-app
VeriTeknik
smart-mcp-proxy