A powerful financial integration skill that enables AI agents to manage direct bank payments, transactions, and budgets via the Banked API.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install banked
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 banked 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 Banked?
The Banked skill for Openclaw Skills provides a robust bridge between AI coding agents and the Banked payment ecosystem. By leveraging the Membrane framework, this skill allows developers to bypass the complexities of traditional card payments and interact directly with bank accounts for a more streamlined financial workflow. It is designed to handle sensitive financial data with high security, ensuring that authentication is managed server-side without exposing local secrets.
This integration is essential for developers building fintech applications or automated accounting bots. Whether you are tracking transaction history, managing organizational budgets, or setting financial goals, this skill simplifies the process by providing high-level actions that abstract away the raw API plumbing. It ensures that your AI agent can focus on logic and decision-making rather than the nuances of payment gateway protocols.
Banked Use Cases
- Automating the retrieval and analysis of bank transaction history for accounting.
- Setting up and monitoring organizational budgets directly through AI commands.
- Integrating direct bank-to-bank payment methods into e-commerce or service platforms.
- Tracking progress toward financial goals and providing real-time updates to users.
- Managing payment workflows without the overhead of traditional credit card processing fees.
How Banked Works
- The AI agent initializes the environment by installing and logging into the Membrane CLI to establish a secure session.
- A connection to Banked is ensured using the domain, which triggers the automated OAuth or API key handshake managed by Membrane.
- The agent polls the connection status until it reaches a READY state, handling any required user actions like manual browser authentication.
- Using natural language intent, the agent searches for specific Banked actions such as listing transactions or managing budgets.
- The agent executes the chosen action with JSON parameters, or performs direct proxy requests to the Banked API for custom use cases not covered by standard actions.
- Membrane handles all credential refreshing and header injection transparently throughout the lifecycle.
Banked Setup
First, install the Membrane CLI to enable Openclaw Skills to interact with the service:
npm install -g @membranehq/cli@latest
Next, authenticate the agent. Use the --clientName flag to specify your environment (e.g., openclaw):
membrane login --tenant --clientName=openclaw
Establish a connection to the Banked service:
membrane connection ensure "https://banked.com/" --json
If the connection state is BUILDING, wait for it to be ready:
npx @membranehq/cli connection get <id> --wait --json
Banked Data Schema & Taxonomy
The skill organizes data around core financial entities. All interactions return structured JSON for easy parsing by AI models.
| Entity | Description |
|---|---|
| Account | Contains details regarding the connected bank account and associated balances. |
| Transaction | Records of individual money movements, including metadata like timestamps and amounts. |
| Budget | Data structures for defining spending limits and tracking current expenditures. |
| Goal | Specific financial targets with progress tracking metrics. |
Actions return an output field containing the processed data or the result of the financial operation.
Banked Advanced Features
- Natural Language Action Discovery: Find the right tools by describing your intent rather than memorizing API endpoints.
- Transparent Proxy Requests: Perform raw API calls while Membrane automatically manages base URLs and authentication headers.
- Long-polling Wait Support: Utilize the
--waitflag for asynchronous connection building, ensuring the agent doesn't fail during setup. - Headless Authentication: Support for server-side or CLI-only environments via authorization URL generation and code completion.
- Multi-Agent Optimization: Fine-tune tooling responses by specifying the agent type (e.g., Claude, Windsurf) during the login process.
SKILL.md
Loading
Related Openclaw Skills
EverWebinar
The EverWebinar skill enables AI agents to manage automated webinar series, user data, and marketing tags through the Membrane CLI.
M3ter Skill for AI Agents
A powerful integration for AI agents to monitor, analyze, and optimize cloud infrastructure spending through the M3ter platform.
Paylocity
A comprehensive HCM integration that enables AI agents to automate payroll, employee management, and HR workflows via a secure CLI-driven interface.
Mslm Cloud
A powerful integration skill for managing Mslm Cloud storage, users, and organizational data through the Membrane framework.
Oracle Cloud HCM Skill
An AI-powered integration skill for managing workforce data and human capital workflows within Oracle Cloud HCM.
Finage
A powerful financial data integration for Openclaw Skills that provides real-time access to global stock, forex, cryptocurrency, and commodity market data.
Featured*
