tables_list
List all tables in your system
LABSCO SUMMARY
A business back office an agent can operate end to end.
It goes past reading data: one session can define the table, install a template's worth of structure, wire an automation and push the result out over WhatsApp. That suits running a small operation from chat, not just reporting on one.
What it is
Runik AI's remote MCP server at https://runikapp.com/mcp. It exposes a no-code business data platform as tools: tables and their records, views such as kanban and calendar, industry templates, automations, and outbound messages over WhatsApp, Telegram and Discord.
What you get
- Tables listed, inspected for their columns, types and relations, and created
- Records listed with filters, search and pagination, then read, created, updated and deleted
- Views — kanban, calendar, chart — listed and created
- Industry templates installed complete with tables, views and sample data
- Automations listed, created from a source, process and destination, and run on demand
- Messages sent out over WhatsApp, Telegram or Discord
Requirements
A Runik account. Authentication is OAuth with PKCE, and the server handles Dynamic Client Registration itself, so there is no key to paste. Streamable HTTP is the recommended transport; an SSE endpoint at https://runikapp.com/mcp/sse remains for older clients.
Setup effort
One command, then sign in — npx @smithery/cli install @shabit/runik-ai, then approve the OAuth sign-in
USE CASES
01
Standing up a system for a new line
Browse the industry templates and install one, arriving with tables, views and sample data already in place.
02
Adding a table without opening the builder
List what already exists to avoid a duplicate, then describe the columns you need and create it.
03
Understanding a table before you query it
Read its columns, types and relations, so the follow-up questions are asked against fields that exist.
04
Giving the team a board to work from
Create a kanban or calendar view over an existing table, so the data has somewhere to be worked rather than only stored.
05
Automating the follow-up
Define the automation's source, processing and destination, then run it on demand instead of waiting for its trigger.
06
Getting the result to a person
Send the outcome straight to WhatsApp, Telegram or Discord rather than copying it out of the workspace.
18 tools
Listed in project docs. Your client may show a different list.
table_schema
Get table structure (columns, types, relations)
table_create
Create a new table with columns
{table}_list
One per table — {table} is the table's name. List records with filters, search, pagination
{table}_get
One per table — {table} is the table's name. Get a single record by ID
{table}_create
One per table — {table} is the table's name. Create a new record
{table}_update
One per table — {table} is the table's name. Update a record
{table}_delete
One per table — {table} is the table's name. Delete a record
views_list
List configured views (kanban, calendar, chart...)
view_create
Create a new view
template_list
List 26 industry templates
template_install
Install a template with tables, views, and sample data
automations_list
List automations
automations_create
Create an automation (source, process, destination)
automations_run
Run an automation
whatsapp_send
Send WhatsApp message
telegram_send
Send Telegram message
discord_send
Send Discord message
BEFORE YOU INSTALL
What to know first
Tools you get
18 tools named in the project's documentation. Five of these names are templates ({table}_list, {table}_get, {table}_create, {table}_update and {table}_delete): each exists once for every table in your workspace, with {table} replaced by that table's name, so the number of tools grows with every table you add.
Requires
A client that can reach a remote MCP server over HTTP
INSTALLATION
Connect your client
Server setup
npx
Smithery
TERMINAL
npx @smithery/cli install @shabit/runik-aiHosted endpoint
ENDPOINT URL
https://runikapp.com/mcpAdd this hosted endpoint to your MCP client, then sign in.
MCP Clients
Cline
CLINE · JSON
{
"mcpServers": {
"runik-ai": {
"url": "https://runikapp.com/mcp",
"type": "streamableHttp"
}
}
}
