obsei

Level ALocal

Open source, privacy-first Voice of Customer analytics. A self-hosted alternative to Enterpret and Unwrap that strips PII from feedback before any LLM sees it.

Overview

obsei is a self-hosted, open source Voice of Customer analytics platform and an alternative to Enterpret and Unwrap. It redacts PII before any LLM sees the data, ingests feedback from reviews, tickets, calls and surveys, and provides sentiment, intent and language classification plus theme clustering. An MCP server and Claude plugin let AI agents answer customer questions with cited, privacy-filtered evidence.

⭐ 1.4kDevOps & DeploymentPythonRepo ↗Updated 2026-11-01

Installation

MCP client config (paste into mcp.json)

{
  "mcpServers": {
    "obsei": {
      "command": "uvx",
      "args": [
        "--from",
        "obsei>=1.0",
        "obsei",
        "mcp"
      ]
    }
  }
}

Agent install prompt

Copy and paste into Cursor / Claude / other agents to install this MCP for you

Please read this MCP repository and install it into my current environment:
https://github.com/obsei/obsei

Known working client config (can write into mcp.json):
{
  "mcpServers": {
    "obsei": {
      "command": "uvx",
      "args": [
        "--from",
        "obsei>=1.0",
        "obsei",
        "mcp"
      ]
    }
  }
}

Please: 1) check README for deps and env vars; 2) write config for my client; 3) if secrets are missing, tell me what to fill — do not invent values.

Related servers