Tutorial | B.AI Skills Minimal Deployment and Operation Guide, Infusing Financial Genes into AI Agents

In today’s era of rapid advancements in large model technology, artificial intelligence is accelerating the reshaping of productivity across thousands of industries. However, in this exciting process, the industry has always faced a key implementation bottleneck: most current AI Agents still remain at the level of “information processing” and “dialogue interaction.” When it comes to rigorous fund transfers, asset management, or complex financial transactions, they are often limited by a lack of specialized execution interfaces, falling into the dilemma of “understanding requirements but unable to take action.”

To bridge the gap from “cognition” to “execution,” AI urgently needs a dedicated underlying financial operation system. This is the original intention behind the birth of B.AI. As a pioneer focused on building AI financial infrastructure, B.AI’s core vision is to lay down a native financial service system and underlying communication channels for countless future AI Agents. To thoroughly break down this execution barrier, the B.AI team has deeply encapsulated complex financial logic on top of industry-standard intelligent agent architectures and open-sourced a Skills module specifically for AI Agent financial operations.

This article will introduce you to the Skills provided by B.AI in an easy-to-understand manner, showing how they endow AI Agents with real financial execution capabilities, and will guide you step-by-step through local deployment and practical invocation.

1. Understanding the Modular Financial Skills Exclusive to AI Agents

Simply put, an AI model itself only possesses powerful text processing and logical generation capabilities, like a “brain” with vast knowledge but lacking limbs. Skills are the “hands, feet,” and “professional toolkits” that empower this brain. Today, loading modular Skills to expand AI’s execution power has become a universal standard across major mainstream intelligent agent frameworks. It includes a series of reusable, task-specific functional instruction sets.

Every Skill offered by B.AI encapsulates highly specialized knowledge in finance and data fields, providing a highly automated unified deployment solution. Through minimal Skills deployment, an AI Agent can perform complex financial operations, payment protocols, and account asset management across platforms based on natural language instructions.

Its core value lies in:

Capability Expansion: Transforming AI from a mere “chat assistant” into a “smart operator” with financial execution power.

Deterministic Execution: Constraining AI’s behavior paths to ensure every transaction and query complies with financial security standards, significantly reducing risks from AI hallucinations.

LEGO-Style Composability: Focusing on providing high-value, general-purpose financial Skills that can seamlessly connect with various mainstream frameworks (such as OpenClaw). Multiple Skills can also work together to form complex financial workflows, truly strengthening the underlying connection between AI and the financial infrastructure.

2. Skills Installation Guide

Thanks to B.AI’s unified installation program, even users without development experience can quickly complete full-chain local deployment, enabling AI Agents to leap from “general dialogue” to “professional financial execution” with a single click.

1. Environment Preparation

Before starting installation, ensure your local development environment is ready:

• OpenClaw or compatible intelligent agent environment.

• Node.js: Since the installation uses npx for one-click setup, Node.js is a must-have platform-wide.

• Python 3 & Git: For scripting and code cloning.

Tips: After running the installation script, the system will automatically perform checks, and users can follow system prompts for configuration.

2. Run the Installation Command

Open the corresponding command-line tool for your OS (macOS / Linux users open Terminal, Windows users use Git Bash), copy and input the following command and press Enter: npx skills add

The system will prompt you to install the base package skills@1.4.9 (version may update). Enter y to confirm, and then the system will automatically clone the resource repository.

3. Select Skills as Needed

After cloning, the terminal will automatically scan and list the 8 core Skills currently supported. You can browse using the up/down arrow keys, select or deselect with the spacebar, and confirm with Enter. The Skills currently provided by B.AI include:

SunSwap DEX Trading: Supports quoting, token swapping, liquidity, and pool operations.

SunPerp Perpetual Futures Trading: Perpetual contract market data queries, order placement, and position management.

TRC20 Token Toolkit: General token metadata queries, balance checks, and transfers.

TronScan Data Lookup: Comprehensive blockchain data, transactions, tokens, and network statistics.

Multi-Sig & Account Permissions: Native multi-signature workflows and account permission management.

recharge-skill & x402-payment: Account recharge and professional payment API resource protocols.

4. Choose Installation Scope

After selecting Skills, the system will ask about your installation scope:

Project (recommended): Install in the current project directory (e.g., ~.agents\skills), making it easy to submit and version control along with your project code.

Global: Install globally, effective for all local projects.

Once selected, the installer will automatically complete subsequent file linking and configuration. When the terminal outputs “Done,” the Skills have been successfully installed.

3. Practical Interaction Guide Based on OpenClaw

After the installation command completes, your AI Agent (using OpenClaw as an example) will have officially loaded this set of professional financial operation Skills. Next, you can command your AI Agent through the OpenClaw interface in three typical scenarios.

Scenario 1: Query Installed Skills (Status Confirmation)

Before executing complex tasks, if you want to confirm whether the tools just installed are active, you can directly ask the AI in natural language for a “status check.” For example, “Please check which Skills are currently loaded in your local directory.” OpenClaw will automatically check and list the deployed core modules, such as sunswap-dex-trading, agent-wallet, tronscan-data-lookup, etc.

Scenario 2: Implicit Triggering (Natural Language Driven Execution)

This is the most intuitive interaction method. Users do not need to mention specific code files, just give instructions as if talking to a human assistant. For example, “Please check the real-time exchange rate of USDT to TRX on SunSwap.” Upon receiving this natural language command, OpenClaw will precisely match and activate the relevant Skill in the background. It will then automatically call the relevant interfaces to complete the query and report the final result directly to you.

Scenario 3: Chained Calls (LEGO-Style Coherent Workflow)

Thanks to the independent yet seamlessly combinable “LEGO” nature of Skills, users can give the AI complex multi-step tasks, allowing it to autonomously invoke different components to complete a closed loop. For example, “First, check how much USDT I have at this address (enter your address), then calculate how much I can get if I swap all my TRX to USDT at the current SunSwap optimal rate.”

Special Tips:

In the above scenarios, we demonstrated the full capabilities of B.AI Skills. But please note: if the instructions involve actual fund transfers, asset swaps, or multi-signature authorizations on the blockchain, you need to use B.AI’s specially developed Agent Wallet component. Agent Wallet is a secure signing environment designed for intelligent agents. When executing financial tasks, it employs an innovative “encrypted file + authorization password” dual-lock mechanism to securely isolate the AI Agent’s operational environment.

To explore more about the underlying architecture and cutting-edge features empowering AI Agents, visit the official B.AI website for an in-depth experience and join us in pioneering the new era of intelligent agent economy.