Knowi

by cloud9charts

257 downloads
Not rated
GitHub

About

Query databases, build dashboards, and analyze documents in natural language from Claude or any MCP client. 28+ tools, 70+ data sources.

Details

Author
cloud9charts
Downloads
257
Categories
Cloud Service, Other, Database

- Built directly into Knowi; no separate server installation
- AI‑powered tools: knowi_do, knowi_ask, knowi_search
- Deterministic tools: list dashboards, get/push data, create queries, widgets, reports, alerts
- Supports file import (CSV, Excel) and document ingestion for AI Q&A
- Token management with one token per user and easy rotation
- Security: scoped credentials, authorization on every request, prompt injection protection, destructive operation safeguards, data truncation

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

  1. Download and install Highlight from highlightai.com/download
  2. Navigate to the plugins tab and select "Add Custom Plugin"
  3. Configure the plugin with the settings below
    Plugin Name Knowi
    Command (node, npx, python, etc.)

    Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.

  4. Enable "Start Automatically" if you want the plugin to start when Highlight launches

From the repository

Generate an MCP token in Knowi (Settings → AI Settings → MCP Token), then configure your AI client. For Claude Code, run claude mcp add knowi https://your-instance.knowi.com/api/2.0/mcp --transport http --header "Authorization: Bearer YOUR_MCP_TOKEN". For Claude Desktop, edit claude_desktop_config.json with the appropriate URL and token, then restart the desktop app. Verify by asking “What Knowi tools are available?”

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "knowi": {
            "knowi": {
                "type": "streamable-http",
                "url": "https://your-instance.knowi.com/api/2.0/mcp",
                "headers": {
                    "Authorization": "Bearer YOUR_MCP_TOKEN"
                }
            }
        }
    }
}

McpServers

{
    "knowi": {
        "type": "streamable-http",
        "url": "https://your-instance.knowi.com/api/2.0/mcp",
        "headers": {
            "Authorization": "Bearer YOUR_MCP_TOKEN"
        }
    }
}

Knowi MCP Server

Connect Claude Code, Claude Desktop, Cursor, or any MCP-compatible AI client to Knowi. Query your data, create dashboards, manage reports, and more — in natural language, directly from your AI assistant.

The MCP server is built into the Knowi application. No separate installation required.

Quick Start

1. Generate an MCP Token

1. Log into Knowi
2. Go to Settings → AI Settings
3. Under MCP Token, choose an expiry and click Generate Token
4. Copy the token — it is shown only once

2. Connect to Claude Code

claude mcp add knowi https://your-instance.knowi.com/api/2.0/mcp \
  --transport http \
  --header "Authorization: Bearer YOUR_MCP_TOKEN"

For knowi.com accounts:

claude mcp add knowi https://www.knowi.com/api/2.0/mcp \
  --transport http \
  --header "Authorization: Bearer YOUR_MCP_TOKEN"

3. Connect to Claude Desktop

Edit your config file:

- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
- Linux: ~/.config/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "knowi": {
      "type": "streamable-http",
      "url": "https://your-instance.knowi.com/api/2.0/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_MCP_TOKEN"
      }
    }
  }
}

Restart Claude Desktop after saving.

4. Verify

"What Knowi tools are available?"

You should see tools including knowi_do, knowi_ask, knowi_search, knowi_list_dashboards, and others.

---

Available Tools

AI-Powered Tools

Natural language tools that use Knowi's AI orchestrator to interpret instructions and chain operations.

| Tool | Description |
|------|-------------|
| knowi_do | Execute any Knowi operation via natural language |
| knowi_ask | Ask a data question — finds the dataset, generates a query, returns results |
| knowi_search | Search across all dashboards, widgets, datasets, and reports |

Deterministic Tools

Direct, fast operations with predictable behavior.

| Tool | Description |
|------|-------------|
| knowi_list_dashboards | List all accessible dashboards |
| knowi_get_data | Retrieve data rows from a widget or dataset |
| knowi_push_data | Push data rows to a dataset (creates it if it doesn't exist) |
| knowi_get_details | Get full metadata for any asset |
| knowi_find_widget | Find widgets by name |
| knowi_create_datasource | Create a new datasource connection |
| knowi_create_query | Create and run a query against a datasource |
| knowi_create_widget | Create a new chart or visualization |
| knowi_update_widget | Update an existing widget |
| knowi_create_report | Create a scheduled report |
| knowi_create_alert | Create a data alert |
| knowi_list_alerts | List configured alerts |
| knowi_test_alert | Test an alert |
| knowi_update_alert | Update an alert |
| knowi_run | Trigger execution of a query or report |
| knowi_export_pdf | Export a dashboard or widget as PDF |
| knowi_export_csv | Export data as CSV |
| knowi_get_embed_url | Generate an embeddable share URL |
| knowi_get_insights | Get AI-generated insights for a widget |
| knowi_layout_dashboard | Arrange widgets on a dashboard |
| knowi_import_file | Import a file (CSV, Excel) as a dataset |
| knowi_ingest_document | Ingest a document for AI Q&A |
| knowi_ask_document | Ask questions against ingested documents |
| knowi_extract_document_data | Extract structured data from a document |
| knowi_list_documents | List ingested documents |
| knowi_delete_document | Delete an ingested document |
| knowi_delete | Delete an asset (requires explicit confirmation) |

---

Usage Examples

"Show me my Knowi dashboards"

"Ask Knowi: what were our top customers by revenue last month?"

"Create a sales dashboard with revenue trends and top products"

"Search Knowi for dashboards related to customer analytics"

"Get the data from the Monthly Revenue widget"

"Export the Executive Summary dashboard to PDF"

"Push these sales records to a Knowi dataset called 'Daily Sales'"

"Set up a daily email report for the executive dashboard"

---

Token Management

Rotate a token:
1. Go to Settings → AI Settings → MCP Token
2. Click Revoke Token
3. Generate a new one and update your config:

claude mcp remove knowi
claude mcp add knowi https://your-instance.knowi.com/api/2.0/mcp \
  --transport http \
  --header "Authorization: Bearer YOUR_NEW_TOKEN"

One token per user. Generating a new token automatically revokes the previous one.

---

API Reference

The MCP server exposes both Streamable HTTP (for MCP clients) and REST endpoints (for direct programmatic use).

Health Check

curl https://your-instance.knowi.com/api/2.0/mcp/health
{
  "status": "healthy",
  "toolCount": 28,
  "tools": ["knowi_do", "knowi_ask", "knowi_search", "..."],
  "activeSessions": 0
}

List Tools

curl -H "Authorization: Bearer YOUR_TOKEN" \
  https://your-instance.knowi.com/api/2.0/mcp/tools

Call a Tool (REST)

curl -X POST https://your-instance.knowi.com/api/2.0/mcp/tools/call \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "knowi_list_dashboards",
    "arguments": {}
  }'

Streamable HTTP (JSON-RPC)

All MCP client operations use POST /api/2.0/mcp with standard JSON-RPC:

```bash

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.