A powerful integration for AI agents to interact with Retool data, manage users, and execute complex database or API queries via the Membrane framework.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install retool
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install retool using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Retool AI Agent Skill?
The Retool skill for Openclaw Skills enables AI agents to seamlessly interact with Retool, a leading low-code platform for building internal tools. By leveraging the Membrane CLI and protocol, this skill allows agents to bypass the complexities of manual authentication and credential management. It provides a standardized interface for executing queries across a vast array of data sources, including SQL databases, REST APIs, GraphQL, and various SaaS platforms like Salesforce, Stripe, and Slack.
Built for developers and operations teams, this skill empowers Openclaw Skills to act as a bridge between your AI agent and your internal dashboards. Whether you need to fetch analytics, update user permissions, or trigger custom workflows, the Retool skill ensures secure, token-efficient communication by utilizing pre-built actions and server-side credential handling.
Retool AI Agent Skill Use Cases
- Automating data retrieval from Retool-connected databases like PostgreSQL, MongoDB, or Snowflake.
- Managing Retool user access and permissions programmatically via AI agent commands.
- Triggering complex internal workflows and API queries directly from a chat interface.
- Syncing data between Retool apps and external services like Slack, Jira, or Zendesk using Openclaw Skills.
- Building and updating custom admin panels or dashboards through natural language instructions.
How Retool AI Agent Skill Works
- The AI agent initializes a connection to Retool using a normalized URL or domain through the Membrane CLI.
- Membrane handles the OAuth or API key handshake, abstracting the authentication lifecycle from the agent.
- The agent searches for specific Retool actions (e.g., SQL queries, API requests) based on natural language intent.
- Once an action is identified, the agent executes it with required parameters, and Membrane processes the request.
- If a pre-built action is unavailable, the agent uses the Membrane proxy to send authenticated raw requests to the Retool API.
Retool AI Agent Skill Setup
Prerequisites
Install the Membrane CLI to enable Openclaw Skills to communicate with the Retool environment:
npm install -g @membranehq/cli@latest
Authentication
Login to your tenant and specify your agent type (e.g., openclaw):
membrane login --tenant --clientName=openclaw
Create Connection
Establish a secure link to your Retool instance:
membrane connection ensure "https://retool.com/" --json
Retool AI Agent Skill Data Schema & Taxonomy
The skill organizes interaction through a structured schema of connections and actions. Data returned from Retool is typically encapsulated in an output field.
| Entity | Description |
|---|---|
| Connection ID | Unique identifier for the authenticated Retool instance. |
| Action ID | Specific identifier for a Retool operation (e.g., sql_query_v1). |
| Input Schema | JSON object defining the required parameters for an action. |
| Output Schema | JSON structure of the data returned by Retool queries. |
| State | The current status of the connection (READY, BUILDING, or ERROR). |
Retool AI Agent Skill Advanced Features
- Multi-source Querying: Support for hundreds of integrations including BigQuery, Redis, and various AWS services.
- Automatic Credential Refresh: Transparent handling of expired tokens without manual intervention.
- Intent-based Discovery: Dynamically search for available Retool actions using natural language via Openclaw Skills.
- Proxy Request Mode: Direct access to the Retool API endpoints with automatic header injection.
- Long-polling Connection Recovery: Built-in wait mechanisms for connections in a building state to ensure high reliability.
SKILL.md
Loading
Related Openclaw Skills
Check Deployment Status
An automated workflow for AI agents to verify the real-world deployment status of GitHub pull requests and specific commits across staging and production environments.
Session Recovery
An AI session management system that optimizes context usage and enables rapid workflow recovery through a structured 3-level documentation strategy.
Flashloan Analysis Skill
A specialized technical skill for analyzing flashloan operations, evaluating DeFi protocol security, and managing on-chain reference data.
Self-Improvement Skill
A recursive learning framework that enables AI agents to log errors, track user corrections, and evolve through persistent project memory.
Ad Campaign Manager
A Telegram-first ads operations assistant designed to manage reporting, budget pacing, and competitor analysis through a boss-to-assistant workflow.
Video Editing Automation
An automated pipeline to transcribe, split, and subtitle talk-style videos into polished content.
Featured*
