
Supabase Adds Perplexity Computer Integration
Supabase announced it is now available as a connector on Perplexity Computer, the company's agent-based automation platform. The integration allows Computer to use Supabase as a persistent data layer, enabling it to read from and write back to existing Postgres tables without requiring developers to build custom integration code.
The move addresses a common friction point in AI-driven workflows: maintaining state across multiple runs. By connecting directly to Supabase's managed Postgres infrastructure, Computer can store and retrieve information between tasks, making it easier to build applications where continuity matters.
How the Integration Works
With Supabase configured as a connector, Computer can query existing tables, insert new records, and update data as part of its automated workflows. Because Supabase exposes a standard Postgres interface, the integration leverages the database's native capabilities rather than adding a custom API layer on top.
Developers can authenticate Computer using Supabase's existing access controls and Row Level Security policies. This means teams can grant the agent access to specific tables or views without opening their entire database, following the same security model they use for other applications.
The integration also reduces the need for intermediary services or glue code to connect AI agents with production data. Instead of passing data through webhooks or third-party storage, Computer writes directly to the same tables used by the rest of the application stack.
Why Persistent State Matters for AI Agents
Most AI agent platforms are stateless by default. They execute a task, return a result, and forget everything. For simple automations, that works fine. But as workflows become more complex, agents need to remember what happened in previous runs, track progress on multi-step tasks, or coordinate with other systems.
Using a database as the source of truth solves this problem. When an agent writes its results to Postgres, those results are immediately available to other parts of the application. If the agent needs to pick up where it left off, it can query the same tables to reconstruct its state. This approach also makes it easier to debug workflows, since every action leaves a trace in the database.
Supabase has been expanding its integrations with AI coding and automation tools over the past year. In May, the company launched a plugin for AI coding agents that bundles its MCP server and agent skills, giving tools like Cursor and Windsurf direct access to Supabase projects. It also released an official ChatGPT app that lets users manage database infrastructure conversationally.
The Perplexity Computer connector is available now. Developers can find setup instructions in the Supabase documentation.
Sources
1 checkedHow we cover tool news: Create With's tool desk drafts these reports with AI from the sources listed above and checks them against those sources before publishing.








