SmartSuite

Read and write SmartSuite work management data from AI agents. Agents browse solutions and tables, then run full record CRUD: listing, retrieving, creating, updating, and deleting rows in any table.

9 actions API key available

SmartSuite website ↗ Platform Documentation ↗ Capability data verified 2026-07-28

A new form submission arrives from the intake form

Get Table reads the target table's field structure so the write maps to real field slugs

List Records filters for an existing record matched on email or order number

Create Record adds the submission as a new row when no match exists

Update Record patches status and enrichment fields on the existing row when a match is found

The owning team's channel gets the new or updated record with a link

Records the agent could not match confidently, two candidates or a malformed key, wait for a person to resolve

The confirmed resolution is written back, and the flow logs the decision

What This Integration Enables

Teams pick SmartSuite because they outgrew the spreadsheet and built themselves a real operations database: solutions grouping tables, tables holding records, every field typed and named by the people who use it. The failure mode of every self-built system of record is the same, it is only as current as the humans feeding it. FlowRunner agents take over the feeding. Because the connector discovers structure at runtime, reading solutions, tables, and field slugs before it writes, the same flow logic adapts to whatever schema your team has actually built. - Sync inbound leads, orders, and form submissions into tables as new records - Read and filter records with structured queries to drive downstream automation - Patch status and fields as work progresses in other tools, without touching unrelated fields - Discover workspace structure and field schemas before building or running an automation - Retire stale and completed records through supervised cleanup sweeps Full CRUD includes the D, and Delete Record in SmartSuite is permanent. That is precisely where FlowRunner's [human-in-the-loop](/concepts/human-in-the-loop) step earns its place in every sweep. Runtime schema discovery is what makes this durable. When your team adds a status option, renames a field, or spins up a new table, the flow does not break silently: agents re-read the structure from Get Table on each run and map against the current slugs. The workspace can keep evolving the way no-code workspaces do, weekly, opportunistically, without anyone maintaining a brittle field-mapping document on the side.

Without FlowRunner

The operations database trails reality Work happens in forms, inboxes, and stores, and the SmartSuite tables that are supposed to track it get updated when someone has time

Status fields go stale A record says in progress for three weeks because updating it is a manual chore in a second tool

Cleanup is deferred forever Completed and abandoned records pile up until every view needs filters just to be usable

With FlowRunner

Tables update at the speed of events Leads, orders, and submissions become records the moment they exist, matched against what is already there

Status reflects the source of truth Agents patch records as work progresses in other tools, so views and dashboards stay honest

Cleanup is a supervised sweep Agents propose deletions with evidence, and a person approves before anything is permanently removed

Use Case Scenarios

Intake that lands as structured data

A [Typeform](/integrations/typeform) intake form collects requests. The agent reads the target table's schema with Get Table, maps each answer to its field slug, and calls Create Record with properly structured values, status fields as structured objects, dates in the shape the schema expects. Submissions the mapping cannot place confidently go to a person with the agent's best guess. The table stays clean because nothing enters it malformed.

The status board that keeps itself honest

Fulfillment happens elsewhere: an order ships, an invoice is paid, a ticket closes. Each event reaches the agent, which finds the matching record with a List Records filter on the order number and patches exactly one field with Update Record. Omitted fields stay untouched, so the agent never clobbers notes a human added. A [Slack](/integrations/slack) message announces meaningful transitions. The board every standup runs on reflects last hour, not last week.

Mirror the pipeline into reporting without exports

On a schedule, the agent runs List Records with a filter for the reporting slice, active records sorted by due date, and mirrors the result into [Google Sheets](/integrations/google-sheets) for the leadership view. Finance gets the numbers in the tool they live in, the team keeps working in SmartSuite, and nobody maintains a CSV export ritual. The same sweep flags records with past-due dates and posts them to the owner.

Human-in-Loop Highlight

Delete Record is the one operation in this connector that cannot be taken back: the record is permanently removed, and with it every field value, comment thread, and linked-record relationship it carried. In a cleanup sweep the target is, by definition, a record the agent believes nobody needs, and the cost of being wrong is deleting the one row that documented why a customer got a discount or how an incident was resolved. So FlowRunner sweeps never delete on their own judgment. The agent assembles the candidate list, each with its record ID, title, last-modified date, and the filter logic that selected it, and posts: "23 records in Closed Deals untouched for 12 months. List attached. Delete which?" The table owner picks, the agent deletes exactly those and logs the confirmation IDs. An agent can be trusted to find what looks disposable; a person decides what actually is.

Agent processes routinely

Detects exception requiring judgment

Clear match Continues automatically

Ambiguous Routes to human via preferred channel

Human decides

Agent resumes with decision

Agent Capabilities

9 actions

Solutions

List Solutions Lists the workspace's solutions, the top-level containers grouping related tables. The discovery step that orients a flow inside the workspace.
Get Solution Retrieves one solution with its name, description, and display metadata.

Tables

List Tables Lists the workspace's tables, optionally filtered to one solution. Each entry carries the ID needed for record operations.
Get Table Retrieves a table's full field structure: every field's slug, label, and type. The schema read that makes correct writes possible, run before any Create Record or Update Record.

Records

List Records Lists a table's records with structured filtering, sorting, and pagination. The query engine behind matching, reporting, and sweep workflows.
Get Record Retrieves one record with all field values keyed by slug. Used to gather full context before an update or a delete decision.
Create Record Creates a record from a field-slug-keyed object, with complex fields passed as structured values per the table schema. The intake write.
Update Record Patches a record: only the fields provided change, everything else is left untouched. The safe default for status syncs and enrichment.
Delete Record Permanently deletes a record. Cannot be undone. The operation this page's human gate exists for.

Frequently Asked Questions

What can FlowRunner do with SmartSuite?

FlowRunner agents can run List Solutions, Get Solution, and List Tables in SmartSuite, plus 6 more actions.

Does connecting SmartSuite to FlowRunner require OAuth?

No. SmartSuite connects to FlowRunner with an API key, no OAuth flow required.

Can SmartSuite trigger a FlowRunner workflow automatically?

SmartSuite doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with SmartSuite

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

Start Building Free Talk to us