--- title: "Supabase Integration" description: "Real-time database operations and event triggers for Supabase. Agents react to record creation, updates, and deletion. Full CRUD with PostgREST filter support and dynamic table discovery." url: https://flowrunner.ai/integrations/supabase date_modified: 2026-08-01T02:40:32-07:00 --- # Supabase [Database](https://flowrunner.ai/integrations/category/databases-warehouses) Real-time database operations and event triggers for Supabase. Agents react to record creation, updates, and deletion. Full CRUD with PostgREST filter support and dynamic table discovery. [Verified](https://flowrunner.ai/integrations/verified "What does verified mean?") · 3 triggers · 4 actions · API key · available [Supabase website](https://supabase.com/) · Capability data verified 2026-06-20 1. On Record Created fires when user completes a key action in the SaaS application 2. Agent reads event type field to determine routing destination 3. Agent routes billing events to Stripe and onboarding completions to HubSpot 4. Agent calls Insert Record to write processed results back to Supabase 5. Agent calls Update Record to set processed\_at timestamp on completed records 6. Operations team notified of sync summary via Slack 7. Records with anomalous field values held for operations review before downstream sync ## What This Integration Enables Supabase integration gives agents bidirectional database access with real-time event triggers. Agents read data to make decisions, write results back to Supabase, and react to data change events that drive downstream workflows. Dynamic table and column discovery means agents adapt to your schema without hardcoded configuration. ### Without FlowRunner **Custom webhook code required**: Every application event connecting to a business system needs a hand-coded integration **No processing status**: No way to tell which records have been processed in a batch workflow **Data quality issues propagate**: Out-of-range values in Supabase sync to ERP and CRM without any check ### With FlowRunner **No-code application event routing**: Any Supabase record change drives operational workflows without code **Idempotent batch processing**: processed\_at timestamps prevent any record from being processed twice **Data validation at the source**: Anomalous records held for review before they reach connected systems ## Use Case Scenarios ### Application Event Processing A SaaS product uses Supabase as its backend database. When a user completes a key action (submitting a form, completing onboarding, triggering a billing event), a record is written to the Supabase events table. On Record Created fires. The agent reads the event type and routes it: billing events go to Stripe, onboarding completions update HubSpot, form submissions trigger the downstream intake workflow. Application events drive operational workflows without custom webhook code. ### Bidirectional Data Sync An agent processes records from an external source (NetSuite, ShipBob, Parseur). After processing, it calls Insert Record to write the structured results back to Supabase, making them available to the application's frontend or other services. The application always has access to current processed data. ### Status Tracking A workflow processes records from Supabase in batches. After processing each record, the agent calls Update Record to set a processed\_at timestamp and a status field. Select Records with a null processed\_at filter identifies unprocessed records for the next batch. The workflow is idempotent: no record gets processed twice, and progress is tracked in the same database. ## Human-in-Loop Highlight When On Record Created or On Record Updated fires for a record with field values that fall outside expected ranges (anomalous amounts, unusual status combinations, unexpected foreign keys), the agent pauses before processing it downstream. It routes to the operations team via Slack with the raw record data: "Anomalous record detected in \[table\]: \[field values\]. Review before I process this into connected systems?" The team reviews the raw data before it propagates. Data quality gates are enforced at the source. 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 7 actions ### Capabilities 7 - **Insert Record** Creates a new row in a Supabase table. Used to write agent-processed data, workflow results, or synchronized records back to the application database. - **Select Records** Retrieves records with filtering conditions and column selection. PostgREST filter operators are supported for precise data retrieval. Used when agents need to query application data before making decisions. - **Update Record** Modifies existing records matching filter criteria. Used to update status fields, sync data changes, or mark records as processed. - **Delete Record** Removes records matching filter criteria. Used in cleanup, archival, and data governance workflows. - **Get Tables Dictionary** Returns a list of available tables in the Supabase project. Used for dynamic table selection in workflows that adapt to schema changes. - **Get Columns Dictionary** Returns the column definitions for a specific table. Used when agents need to understand schema structure before inserting or querying records. - **Get Operators Dictionary** Returns the supported PostgREST filter operators available for record queries. Used in dynamic filter construction. ## Triggers 3 triggers ### Event Triggers 3 - **On Record Created** Fires when a new row is inserted into a monitored Supabase table. Used to trigger processing workflows immediately when application data is created: a new order, a new user submission, a new event log entry. - **On Record Updated** Fires when an existing record's data changes. Used to synchronize updates across connected systems or trigger re-processing when application data evolves. - **On Record Deleted** Fires when a record is removed from a table. Used for audit logging, cross-system synchronization, and cleanup workflows. ## Frequently Asked Questions ### What can FlowRunner do with Supabase? FlowRunner agents can run Select Records, Insert Record, and Update Record in Supabase, plus 1 more action. ### Does connecting Supabase to FlowRunner require OAuth? No. Supabase connects to FlowRunner with an API key, no OAuth flow required. ### Can Supabase trigger a FlowRunner workflow automatically? Yes. Supabase supports 3 triggers that can start a FlowRunner workflow automatically. **Work at Supabase?** This integration exposes Supabase 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/supabase. Site index: https://flowrunner.ai/llms.txt