Agentic Trading with Safe Guardrails
AIThis post was created with the assistance of artificial intelligence (AI).

TL;DR

Prime Big Deal Days · Oct 6–7Offer from Amazon

Get the latest gadgets delivered free — and shop member deals

  • Fast, free delivery on millions of items
  • Access to Prime Big Deal Days deals on October 6–7
  • Prime Video, Amazon Music and more included
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

Shuriken has introduced new infrastructure allowing AI agents to perform autonomous trading across various asset classes with built-in permission controls. This development marks a significant advance in autonomous finance, though some details about implementation and security are still emerging.

Shuriken has launched infrastructure that enables AI agents to perform autonomous trading across all major asset classes with built-in safety guardrails, marking a significant step toward autonomous finance.

The platform, called Shuriken, supports trading in on-chain tokens, perpetuals, real-world assets, pre-IPO equity, and prediction markets, operating 24/7. It allows users to connect their agents to various information sources, such as Twitter, Telegram, Discord, and on-chain activity, enabling agents to act proactively before market moves.

Shuriken emphasizes that agents only perform actions permitted by the user, with granular permission controls that eliminate the need for seed phrases in runtime. The platform provides integration guidance for developers, including how to authenticate, scope permissions, and call APIs or SDKs. The system uses a Rust crate called shuriken-skills, which packages guidance skills for agent reasoning about integration and permissions.

The infrastructure includes native plugin manifests for popular coding agents like Claude, Codex, Copilot, Gemini, and Cursor, allowing seamless installation via their respective plugin systems. The release also details how external developers can publish and update skills on various marketplaces, with a focus on security and permission scoping.

Why It Matters

This development is significant because it introduces a framework for autonomous trading agents to operate safely across multiple asset classes, potentially transforming financial automation. By integrating granular permissions and safety guardrails, Shuriken aims to mitigate risks associated with autonomous trading, which has been a concern in the evolving AI-driven finance sector. The platform could accelerate the adoption of autonomous finance tools, but also raises questions about security, oversight, and regulatory compliance.

Amazon

autonomous trading software

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background

Autonomous agents have been capable of many tasks but were limited in trading due to safety and permission concerns. Shuriken’s infrastructure builds on ongoing efforts to enable AI-driven trading while maintaining user control. Prior developments in autonomous finance have focused on simulations or limited environments; this release marks a move toward real-world, cross-asset trading with safety features.

The platform’s announcement follows broader industry interest in AI-powered financial tools, with some skepticism about risk management and security. The release of detailed integration guidance and marketplace tools suggests a push toward wider adoption and developer engagement in autonomous trading systems.

“We are at the beginning of autonomous finance. Our platform allows agents to trade safely across major assets, with permissions and guardrails that protect users.”

— Shuriken team

“Our integration guidance helps developers build secure, permissioned agents that can operate in real-time markets without risking user funds or privacy.”

— Shuriken developer

Amazon

AI trading agent platform

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

What Remains Unclear

It is not yet clear how widely adopted the platform will become or how effective the guardrails will be in preventing unintended trading behaviors. Details about security audits, regulatory compliance, and real-world performance are still emerging.

Amazon

permission-controlled trading bots

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

What’s Next

Next steps include wider developer adoption, integration of additional asset classes, and potential regulatory review. Monitoring real-world deployment and security performance will be key in assessing the platform’s impact.

Amazon

blockchain asset trading tools

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

How does Shuriken ensure the safety of autonomous trading agents?

Shuriken provides granular permission controls and guardrails that restrict agent actions based on user-set scopes, reducing risks of unwanted trades or security breaches.

Can individual users implement their own trading agents on Shuriken?

Yes, the platform offers integration guidance and plugin support for developers to build and deploy custom agents within the safety framework.

What assets can agents trade using Shuriken?

Agents can trade on-chain tokens, perpetuals, real-world assets, pre-IPO equity, and prediction markets, according to the platform’s current support.

Are there any security or regulatory concerns with autonomous trading?

While Shuriken emphasizes safety guardrails, the broader risks of autonomous trading, including security breaches and regulatory compliance, remain areas to watch as the platform develops.

HALLOWEEN

Halloween Picks

As an affiliate, we earn on qualifying purchases.

You May Also Like

Japan’s Sumitomo Metal, Sojitz look to Southeast Asia for rare earths

Japanese firms Sumitomo Metal Mining and Sojitz are developing rare-earth resources in Southeast Asia to establish China-independent supply chains, with projects in the Philippines, Vietnam, and Malaysia.

Measuring Input Latency On Linux: X11 Vs. Wayland, VRR, And DXVK

New tests compare input latency on Linux between X11 and Wayland, with focus on VRR and DXVK effects, revealing significant performance differences.

What A $325M Bribery Case In China Signals For Global Trade

IdeaNavigator AI flags a report of a Chinese official sentenced over $325 million in bribes, but provides no case details to verify trade implications.

Transforming Manufacturing Onboarding With Cutting-Edge VR Solutions

Factories are testing VR training modules to speed onboarding, reduce safety risks, and address labor shortages with new immersive tech.