--- title: "SeaTable Integration" description: "Connect AI agents to a SeaTable base. Agents read, append, and update rows in bulk, run read-only SQL over base data, maintain row links, and discover table and column names." url: https://flowrunner.ai/integrations/seatable date_modified: 2026-08-01T02:40:32-07:00 --- # SeaTable [Database](https://flowrunner.ai/integrations/category/databases-warehouses) Connect AI agents to a SeaTable base. Agents read, append, and update rows in bulk, run read-only SQL over base data, maintain row links, and discover table and column names. [Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 13 actions · API key · available [SeaTable website](https://seatable.io/) · [Platform Documentation](https://api.seatable.io/reference/) · Capability data verified 2026-07-14 1. A batch of new records arrives from another system 2. Agent reads the records and maps them to SeaTable column names 3. Agent runs Append Rows to bulk-sync up to 1000 rows in one request 4. Agent runs Query with SQL to find rows that need follow-up 5. Agent runs Update Row to record the outcome back on each matching row 6. Agent sends a digest to the team via Slack 7. Any Delete Rows above a threshold pauses for approval ## What This Integration Enables Agents sync records in bulk with Append Rows and Update Rows (up to 1000 per request), log form submissions and webhook payloads as new rows, query and aggregate base data on demand with read-only SQL, maintain relationships between tables with row links, and discover table and column names before performing data operations. Query with SQL runs read-only SELECT statements against base tables and converts result column keys to human-readable names, so the output is easy to route onward. Get Base Metadata resolves the table and column names an agent needs before it writes. ### Without FlowRunner **Row-by-row imports**: Loading a batch means many single-row appends **Reporting outside the base**: Data is exported elsewhere to run a summary query **Bulk deletes unguarded**: A batch delete has no checkpoint against mistakes ### With FlowRunner **Bulk appends**: Append Rows syncs up to 1000 rows in a single request **SQL in the base**: Query with SQL aggregates base data directly, with readable columns **Guarded deletes**: Delete Rows above a threshold pauses for human approval ## Use Case Scenarios ### Bulk sync into a base New records arrive, for example rows from a spreadsheet. The agent reads them and calls Append Rows to bulk-sync up to 1000 at a time, so a large import is a handful of requests rather than hundreds. The team's base stays current with the source system. ### Report and write back The agent calls Query with SQL to pull rows matching a condition, sends a digest to the team with Slack, then calls Update Row to record the outcome (for example a created reference in another tool) back on each matching row. Reporting and write-back happen in one flow. ### Cleanup with a human gate An agent is asked to purge rows. Before it runs Delete Rows above the configured threshold, it does not act on its own. It routes the affected rows and a count for approval and deletes only after a person confirms. ## Human-in-Loop Highlight Appends and updates in SeaTable are safe to repeat; a bulk Delete Rows is the one that removes data collaborators may still rely on. 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 would remove more rows than the configured threshold, the agent pauses before executing and asks through Slack: "This Delete Rows on the `Projects` table would remove 180 rows. Here is a sample. Approve, narrow the set, 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 13 actions ### Metadata 1 - **Get Base Metadata** Returns the tables and columns of the base, so an agent can resolve names before reading or writing. ### Rows 8 - **List Rows** Reads rows from a table, optionally through a view, with start and limit pagination up to 1000 rows. - **Get Row** Fetches one row by id. - **Append Row** Appends one row to a table, keyed by column name. - **Append Rows** Appends many rows in one request (up to 1000), for bulk syncs. - **Update Row** Updates one row by id. - **Update Rows** Updates many rows in one request. - **Delete Row** Deletes one row by id. - **Delete Rows** Deletes many rows in one request. ### SQL 1 - **Query with SQL** Runs a read-only SELECT against base tables, with human-readable result column names. ### Links 3 - **Add Link** Links two rows across tables. - **List Row Links** Lists the rows linked to a row. - **Remove Link** Removes a link between two rows. ## Frequently Asked Questions ### What can FlowRunner do with SeaTable? FlowRunner agents can run Get Base Metadata, List Rows, and Get Row in SeaTable, plus 10 more actions. ### Does connecting SeaTable to FlowRunner require OAuth? No. SeaTable connects to FlowRunner with an API key, no OAuth flow required. ### Can SeaTable trigger a FlowRunner workflow automatically? SeaTable doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger. **Work at SeaTable?** This integration exposes SeaTable to AI agents on every FlowRunner plan, including through MCP, at no cost to you. [See what FlowRunner offers integration partners](https://flowrunner.ai/integrations/partners), including how to keep this page current. --- Markdown version of https://flowrunner.ai/integrations/seatable. Site index: https://flowrunner.ai/llms.txt