Over the past week, we introduced Bitwave Agentic, an expanding collection of infrastructure, interfaces, data, and open-source tools designed for a world in which AI agents increasingly participate in financial operations.
Each announcement addressed a different part of that future. Together, they provide a foundation for agents to work with financial systems while giving finance, accounting, and IT teams the visibility and control those workflows require.
Here’s what we introduced during Bitwave AI Week.
KubeClaw: Controlled infrastructure for autonomous agents
We started the week with KubeClaw, an open-source Kubernetes operator for deploying and managing AI agents in controlled environments.
As agents take on more operational work, businesses need ways to govern where those agents run, which tools they can access, and how they interact with sensitive financial systems. KubeClaw gives technical teams an infrastructure layer for running agents within their own Kubernetes environments while maintaining the controls required by enterprise and regulated organizations.
Agents running through KubeClaw can use tools such as command-line interfaces to perform automated, recurring financial work,including day-over-day accounting workflows in Bitwave.
Learn more at Kube-Claw.com.
Bitwave CLI: An accounting interface built for agents
On Tuesday, we introduced the Bitwave CLI, a command-line interface designed to make accounting work accessible to both AI agents and human operators.
Through the CLI, agents can interact with Bitwave to categorize transactions, generate inventory views, and perform other digital asset accounting workflows. But its capabilities extend beyond digital assets.
Agents can also use the CLI to:
- Create and manage charts of accounts
- Prepare journal entries
- Generate balance sheets and profit-and-loss statements
- Model accounting scenarios
- Create and submit agentic expense reports
- Share completed accounting work through Bitwave’s hosted interface
That final use case points toward a particularly important question: what happens when AI agents begin spending money on behalf of a business?
Just like any employee or operating system incurring an expense, an agent’s financial activity must be documented, categorized, approved, and recorded. With the Bitwave CLI, an agent can prepare an expense report as it works and send the completed accounting into systems such as QuickBooks.
Bitwave MCP: Connecting agents directly to Bitwave
On Wednesday, we released the Bitwave MCP, another interface through which AI agents can work with Bitwave.
Using the Model Context Protocol, organizations can give compatible agents structured access to Bitwave tools and information. An agent can then help categorize transactions, retrieve information, run reports, and complete other accounting workflows through natural-language requests.
Where the CLI provides a command-line interface optimized for agents and developers, the MCP makes Bitwave capabilities available within a growing ecosystem of MCP-compatible AI tools.
Learn more and find setup instructions at Bitwave.io/mcp.
Bitwave Data and Analytics: Eight years of blockchain data, ready for agents
Bitwave has spent the past eight years building the data infrastructure required to support enterprise digital asset accounting. On Thursday, we opened new ways for businesses—and their agents—to work with those datasets through Bitwave Data and Analytics.
Blockchain data can now be queried through the Bitwave CLI, MCP server, REST endpoints, and the Bitwave application itself.
This gives agents the ability to retrieve blockchain data, write and execute SQL-style queries, perform calculations, and create reports without requiring users to manually construct every query. For example, a user could ask an agent to calculate the amount of USDC in circulation, and the agent could translate that request into a query against Bitwave’s datasets.
The result is a powerful research and analytics interface that combines natural-language interaction with the depth of Bitwave’s blockchain data.
Open-source building blocks for what comes next
We also open-sourced libraries that help developers begin exploring two emerging requirements of agentic finance: accounting for agent activity and enabling agents to work with wallets.
The future shape of agentic commerce is still being determined. We don’t yet know exactly how frequently agents will spend money, hold budgets, interact with wallets, or transact on behalf of businesses. But if agents are going to participate meaningfully in economic activity, they will need both the ability to transact and the infrastructure required to account for those transactions.
Our open-source libraries give developers practical building blocks for experimenting with those workflows today.
You can explore Bitwave’s open-source projects on the Bitwave GitHub.
The foundation for agentic finance
AI agents are moving beyond answering questions and generating content. They are beginning to use tools, execute workflows, access data, and participate directly in business operations.
Financial infrastructure must evolve with them.
Bitwave Agentic is designed to connect these emerging systems: controlled agent infrastructure through KubeClaw, accounting capabilities through the Bitwave CLI, tool access through the Bitwave MCP, blockchain intelligence through Bitwave Data and Analytics, and open-source libraries developers can use to explore entirely new financial workflows.
Bitwave AI Week may be over, but this is only the beginning. We’ll be releasing more tools, capabilities, and open-source projects around Bitwave Agentic in the weeks ahead.
Watch the demo of Bitwave AI Week features below, or request early access to features within Bitwave.
Disclaimer: The information provided in this blog post is for general informational purposes only and should not be construed as tax, accounting, or financial advice. The content is not intended to address the specific needs of any individual or organization, and readers are encouraged to consult with a qualified tax, accounting, or financial professional before making any decisions based on the information provided. The author and the publisher of this blog post disclaim any liability, loss, or risk incurred as a consequence, directly or indirectly, of the use or application of any of the contents herein.