Stackby

Connect AI agents to Stackby, the spreadsheet-database. Agents create, read, update, and delete rows in a Stack table as governed steps in a flow.

5 actions available

An inbound record (lead, order, form submission) arrives upstream

Agent reads the record and maps it to Stackby column names

Agent runs Create Rows to log it into the Stack table

Agent runs Get Row to read a status field on a related record

Agent runs Update Rows to advance the record's status

Agent notifies the team with the row details via Slack

Any Delete Rows against a shared table pauses for approval

What This Integration Enables

Agents log inbound records into a Stackby table, sync rows between Stackby and other apps on a schedule or trigger, read a row to check a status field and drive downstream automation, and bulk-create, update, or delete rows from a workflow. List Rows pages through larger tables with Max Records and Offset (up to 100 rows per request), while Create, Update, and Delete each accept up to ten rows per request. Because the service covers row operations only, the Stack id and table name are supplied as plain text.

Without FlowRunner

Manual intake logging Inbound records are typed into Stackby by hand

Status checked by eye Someone opens the Stack to read a status before acting

Deletes without review Removing rows from a shared table has no checkpoint

With FlowRunner

Automated logging Create Rows logs inbound records as a flow step

Status drives the flow Get Row reads a status field and branches the workflow

Guarded deletes Delete Rows against a shared table pauses for human approval

Use Case Scenarios

Intake logging

An inbound record arrives from a form or another connector. The agent maps it to the table's column names and calls Create Rows to log it into the Stack. The intake list that used to be maintained by hand fills itself as records come in.

Status-driven routing

The agent calls Get Row to read a status field on a related record, and when the status warrants action, it branches: it might notify the team with Slack, or call Update Rows to advance the record. The Stack becomes a small state machine the flow can read and write.

Cleanup with a human gate

An agent is asked to remove rows from a shared Stack table. Before it runs Delete Rows, it does not act on its own. It routes the row ids and details for approval and deletes only after a person confirms.

Human-in-Loop Highlight

A Stack table is often shared across a team, so a delete that looked routine can erase records someone else was working from. FlowRunner's answer is human-in-the-loop, an execution pattern where the agent pauses on its own, assembles the context and the choices, routes to a human on their preferred channel, and resumes the moment they respond. When a Delete Rows targets a table flagged as shared, the agent pauses before executing and asks through Slack: "Ready to Delete Rows (3 rows) from `Leads`. Here are the rows. Approve or cancel?" The delete runs only after a person confirms, with the approver and timestamp captured in the run log. A connector can delete any rows; an orchestration layer knows which deletes should stop and ask.

Agent processes routinely

Detects exception requiring judgment

Clear match Continues automatically

Ambiguous Routes to human via Slack

Human decides

Agent resumes with decision

Agent Capabilities

5 actions

Rows

Create Rows Creates up to ten rows in a Stack table from field-keyed objects.
Delete Rows Deletes up to ten rows by id.
Get Row Fetches a single row by id.
List Rows Reads rows from a table with Max Records and Offset pagination (up to 100 per request).
Update Rows Updates up to ten rows, each identified by id.

Start building with Stackby

$100 in credits. No card required. Connect in minutes.

Start Building Free Talk to us