AgentX
Your AI agent thinks. AgentX acts.
AI agents can already reason about your finances. Executing is the harder part. It takes verified authority to act, and a licensed institution to move the money.
AgentX closes that gap. It packages MetaComp’s compliance and regulated financial services, including payments, as Skills your AI agent can call directly, across the LLM platforms you already use. Every action is governed by the StableX Know Your Agent (KYA) Framework, so execution stays compliant, controlled and auditable.
Governed by StableX Know Your Agent (KYA) Framework
Every agent operating through AgentX does so under the StableX Know Your Agent (KYA) Framework, MetaComp’s governance architecture for AI agents in regulated financial services. StableX KYA was launched in April 2026, developed drawing on IMDA’s Model AI Governance Framework for Agentic AI.
Agents are identified before they act, bounded in what they may do, monitored against their mandate, and accountable for how they interact.
01
Agent Identity & Registration — Every agent is known before it acts, bound to an accountable owner with verifiable credentials.
02
Authority & Permission Control — What an agent may do, and to what limit, with human approval where mandates end.
03
VisionX Behaviour Monitoring & Risk Intelligence — What an agent actually did, continuously monitored against its mandate.
04
Ecosystem & Interaction Governance — How agents transact with each other, through mutual verification and shared standards.
The Gaps AgentX Closes
| The Gap | What AgentX does |
|---|---|
| Non-compliant agent actions put capital at risk | Gives your AI agent a compliant path to regulated financial services |
| Digital asset payments carry counterparty risk you cannot see | VisionX Agentic KYT screens wallets and transactions before the money moves |
| Every institution wants a different deployment | Run AgentX yourself, use it out of the box, or give your existing agent Skill-level access |
From Human-Initiated to Agent-Initiated
Financial services have moved from human-to-human, to human-to-app, to human-to-API, to human-to-agent. Agent-to-agent is next with financial transactions initiated and settled between agents. AgentX is built for that transition, and governed from the outset by StableX KYA.
What Is AgentX
AgentX is MetaComp’s agentic financial solution. It packages financial services as Skills that your AI agent can call directly, with every action governed by StableX KYA. Skills run across the LLM platforms you already use, including Claude, Claude Code and OpenClaw, via the Model Context Protocol. Your agent can screen a counterparty with VisionX Agentic KYT before moving on to payments, all in a single platform.
Available now: VisionX Agentic KYT Skill
Screen any wallet or transaction from inside your AI conversation. Paste an address or transaction hash and ask. Your agent returns counterparty risk analysis, sanctions and high-risk exposure screening, and incoming and outgoing exposure distribution.
VisionX Agentic KYT is the first Skill in the AgentX ecosystem, drawing on the VisionX Engine’s on-chain screening. It runs more than four blockchain analytics vendors in parallel, because relying on a single screening tool can leave up to 25% of high-risk exposures undetected. Read the research →
Get started
1
In Claude.ai, open Settings → Connectors → Add custom connector and enter https://www.metacomp.ai/mcp
2Authorise with your MetaComp API key
3Load the Skill from the developer portal (Claude Code only)
4Start checking wallets and transactions
Agentic Skills
Agentic Compliance
Agentic Payments & Wallet
Fiat payment and collection accounts · Stablecoin wallets · Fiat and stablecoin payments · Corporate card and VCC flows (coming soon)
Agentic Wealth¹
Web2.5 agentic asset allocation · Intelligent structured products
AgentX Skills Hub
More Skills from MetaComp and our partners, extending what your agent can do across financial services.
Three ways to deploy
AgentX Platform
White label
Integration
Cloud — deployed through MetaComp’s cloud partners.
SDK — run locally on your own infrastructure, for institutions with data residency or control requirements.
Built for humans and for AI
For humans
Cognitive financial
agents
- Customised and controlled
- Institutional-grade decisioning
- Compliance-integrated
For AI
Autonomous execution infrastructure
- Governed by StableX KYA
- Protocol-agnostic
- Agent-to-agent ready
