A .NET-native MCP orchestrator that sits between an agent and multiple MCP servers, loading downstream tool manifests on demand to reduce always-on context usage. It can also serve agent skills from configured sources and hot-reload configuration changes.
From the registry: Route one agent through one MCP server to many, with progressive tool discovery to shrink context.

Please install the `dotnet-mcp-orchestrator` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install .NET SDK** — The README states the profiling and tool execution commands require the .NET SDK. (https://dotnet.microsoft.com/)
Canonical MCP server config (stdio transport):
- command: `dotnet`
- args: ["tool","execute","McpOrchestrator"]
Note: The simplest portable launch path shown is via the .NET tool executor: `dotnet tool execute McpOrchestrator`. The README's `init` command rewrites an existing MCP host config to use only the orchestrator, but the canonical MCP server command shape is the server launch itself. No required environment variables are shown in this README excerpt. The orchestrator depends on a separate `orchestrator.config.json` for downstream server/skill definitions, generated by `init` or created manually.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Meta Ads MCP (Facebook + Instagram) - analyze performance, manage budgets, pause campaigns.