How to Configure Windsurf Redshift Read Only Access
You set up windsurf redshift read only access by creating a restricted database role. You then pass that role to the editor through an MCP server or shell. Many teams start by giving their coding agent their own personal database login. When a role is too broad, an agent can change or delete warehouse tables. A scoped read role stops the agent from modifying or dropping production data. How windsurf redshift read only access works in practice On June 2, 2026, Cognition renamed Windsurf to Dev
Your coding agents run with your access.
Claude Code reaches prod with a developer's credentials, and approval prompts get clicked through. hoop.dev checks every query, masks sensitive data, and holds risky changes for a human.
You set up windsurf redshift read only access by creating a restricted database role. You then pass that role to the editor through an MCP server or shell. Many teams start by giving their coding agent their own personal database login. When a role is too broad, an agent can change or delete warehouse tables. A scoped read role stops the agent from modifying or dropping production data.
How windsurf redshift read only access works in practice
On June 2, 2026, Cognition renamed Windsurf to Devin Desktop with the same features. The editor and its Cascade agent reach external databases in two main ways. First, the agent can run terminal commands on its own with user permission. It uses a dedicated zsh terminal on macOS or bash terminal on Linux. That terminal loads your shell startup files, so your aliases and variables exist there. If your shell holds database keys, the agent can use them in terminal commands.
Second, the legacy Cascade agent connects to local or remote MCP servers. You define these servers in the editor settings file on your workstation. The MCP server reads your database login details and runs queries for the agent. To keep the database safe, admins create a dedicated database user first. They grant this user usage rights on the target schema and select rights only. They leave out insert, update, delete, and drop rights on every production table. Finally, the developer puts those read credentials into the MCP settings or shell profile.
Why database roles alone leave a gap
A read role is a vital first step for any analytics warehouse setup. It stops the coding agent from dropping tables or changing rows by mistake. It also blocks the agent from creating new objects in your production schemas. Database admins rely on these permissions to protect data integrity across the team.
Still, native database roles only check who logs in and which tables they read. They do not inspect what individual queries do while they run in flight. A read role still lets an agent run a massive join across billion-row tables. That single bad query can lock up compute resources and slow down other users. Even worse, select rights let the agent read every column in an allowed table. If a table holds customer emails, phone numbers, card numbers, or national IDs, queries return them. The raw text flows straight into the agent context window and model prompts. Local shell variables also leave static database passwords sitting in plain text on laptops.
Auditor Read-Only Access + Customer Support Access to Production: architecture patterns and best practices
Continue reading? Get the full guide.
Free. No spam. Unsubscribe anytime.
Closing the gap for AI database queries
Safe agent access requires controls that work at the query and column level. Teams need a way to keep static database keys off developer laptops entirely. Instead of storing long-lived passwords in shell files, sessions should tie to user identity. Every query from an editor session should map back to a specific person.
Data teams also need real-time data masking on query results before agents read them. Masking rewrites matched fields like emails, phone numbers, and card numbers in query results. The agent gets valid schema shapes and sample rows without pulling raw sensitive values. Statement rules can also block risky queries or cap row limits before execution. This keeps warehouse compute stable while giving the coding agent the context it needs.
Add runtime guardrails and masking beside your databases and APIs
Login rules, role grants, and CI checks govern who connects to your stack. Yet any allowed script, engineer, or AI agent still runs commands with the full scope of its role. For workloads on Postgres, MySQL, SQL Server, MongoDB, ClickHouse, Kubernetes, SSH, or internal HTTP services, hoop.dev sits inline in the protocol beside the service as an open-source proxy, a sidecar for your database or API. On every production query an agent or human runs, hoop.dev decides whether to pass, mask, block, or escalate it for review.
Before a query runs, hoop.dev guardrails check the statement against your rules, an AI Analyzer you configure can score intent or hold risky calls for human approval, and hoop.dev data masking rewrites sensitive fields in responses in memory. Explore the MIT-licensed project on GitHub.
Frequently asked questions
Does Windsurf still exist after the Devin Desktop rename?
On June 2, 2026, Cognition renamed Windsurf to Devin Desktop. It keeps the same IDE features, terminal execution, and Cascade MCP support.
Can the coding agent read my local shell database variables?
Yes, the agent runs commands in a dedicated zsh or bash terminal with user permission. That terminal loads your shell profile, making your aliases and exported variables available.
Does a read role stop sensitive data from reaching the AI model?
No, standard select permissions return all columns in a table to the caller by default. You need result masking to rewrite emails, phone numbers, or card numbers in results.
Related guides
In this post