Sabli

app.sabliv1.0.0Updated Oct 4, 2026

Ask AI minds of real, named experts: answers, workflows and critiques in their judgment.

VerifiedStreamable HTTPWeb executableAI & ML

Overview

AI-generated overview

Lets an assistant query AI personas modeled on real, named experts for answers, workflows, and critiques.

What it does
This remote MCP server connects an assistant to AI "minds" based on real, named experts. According to the registry description, it returns answers, workflows, and critiques framed in those experts' judgment. No individual tools are listed in the manifest, so the exact operations are not documented here.
When to use it
Consider it when you want an assistant to answer or critique something from a specific expert's perspective, or to follow a named expert's workflow. It is less relevant if you need general-purpose model access rather than expert-style personas.
Requirements
It runs as a remote streamable HTTP endpoint at sabli.app, so no local runtime or package is needed. The manifest declares no authentication, environment variables, or headers, but the provider's connect page may still require an account or sign-in.
Before you install
The manifest declares no credentials, but the endpoint is third-party and hosted, so anything you send is processed by that provider. Treat the expert personas as AI-generated rather than the real people's own statements, and verify any advice before acting on it.

Installation

In SourceWeft

  1. Open Sabli in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "minds": {
      "type": "http",
      "url": "https://sabli.app/api/mcp"
    }
  }
}

Tools

0
Tool metadata has not been indexed yet.

Version history

1
  1. v1.0.0LatestOct 4, 2026