X-Agent AI
Web3 no-code AI agent creation and operation platform
Performances
RD Popularity Index
RD Growth Index
Comparison
Fundraising
Follow Updates
Follow Lists
People
About X-Agent AI
X-Agent is a no-code AI agent platform that lets users create and deploy AI agents through natural language, with support for Telegram, WhatsApp, and web interfaces. Its core technology includes "Speak to Build," which turns plain-language descriptions into functional agents within minutes, and a Secure Runtime Environment (SRE) that physically isolates private keys while enabling autonomous asset operations and on-chain collaboration.
The platform addresses two key market pains: the technical barrier of building AI agents, and the security risks of entrusting agents with sensitive operations. By removing code requirements and providing a sandboxed execution layer, X-Agent makes agent deployment accessible to non-programmers while maintaining verifiable and auditable actions.
In the past six months (March to September 2026), X-Agent has focused on ecosystem development. It participated in the Open AGI Developer Day in Shanghai with partners including Sentient, Alibaba, and Fudan University, where the team emphasized building a complete chain from agent creation to distribution and monetization. The platform has accumulated over 1 million users who have completed more than 1.1 million agent tasks, consuming 84 billion model tokens. The Litepaper V1 was released during this period, detailing the SRE architecture and the token's role as gas for computation and transaction fees.
No significant negative events were reported in the past two years. The project is distinct from other entities using similar names, as it specifically focuses on the agent application layer with integrated wallet and payment capabilities.
In April 2025, the founder began building xAgent, evolving from a single agent to a multi-agent collaboration system and tackling token cost issues.
In March 2026, the first public version 0.0.4.beta was released; subsequent 0.0.10.beta reduced the base prompt from ~20K to ~5K tokens, significantly improving task stability.
In June 2026, at the Open AGI Developer Day, the company proposed building the Agent Application Layer, aiming to connect creation, deployment, execution, distribution, and monetization.
In July 2026, xpense was open-sourced, creating a payment control plane for AI agents, enabling governed and auditable autonomous payments with OKX Agentic Wallet integration.
In September 2026, Litepaper V1 was released, disclosing no-code deployment and SRE cryptographic sandbox; cumulative users surpassed 1,000,000, with agents completing over 1,100,000 tasks.
Over the past six months (2026-03-01 to 2026-09-01), X-Agent has made several key advances. In July 2026, it officially released Litepaper V1, detailing its no-code deployment and secure sandbox technology for building a decentralized AI agent economy. The project also open-sourced xpense, a payment control plane for autonomous AI agents, integrating with OKX Agentic Wallet and enabling machine-to-machine microtransactions via x402/EIP-3009. Additionally, X-Agent became a merchant on OKX Agent Marketplace, launching three production-grade MCP tools (Token Security Scan, Wallet Reputation, Portfolio Health) for agent-to-agent paid services. Looking ahead, the official roadmap includes opening a no-code public portal in Q3 2026 and launching a decentralized Agent Store in Q1 2027, aiming to complete the full lifecycle from agent creation to monetization.
Ye X. (Co-founder, Head of Product & Technology): LinkedIn profile shows experience at Meta (Technical Program Manager), Lalamove (Senior Program Manager), WeWork (Project Management Lead), J.P. Morgan (Associate Project Manager), and NYU Tandon (MATLAB Tutor). Education: MS from Columbia University, BS from NYU. Twitter: @Ye233618.
Mark Chen (Co-founder, Global Business Lead): LinkedIn profile lists roles at X-Agent (current), with prior experience in business development and partnerships. Education details not fully public. Twitter: @ChenNaiwei.