> For the complete documentation index, see [llms.txt](https://docs.vishwalab.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.vishwalab.com/overview.md).

# Overview

Vishwa is a pre-execution verification control layer for agent finance. It sits between what an AI agent decides and what actually executes on-chain, so funds can move autonomously without handing the agent blind trust.

Every action an agent proposes is submitted as an unsigned request through the Vishwa gateway and checked against the owner's constraints inside a hardware-sealed control layer - a Trusted Execution Environment (TEE) - before anything signs or settles. The agent never holds keys. If a request falls outside the owner's policy, it is denied before execution, not flagged after.

To date, Vishwa's control layer has verified more than $170M in capital allocated and managed under its guardrails.

### The model in one line

The owner defines the rules. The agent operates within them. The control layer enforces them.

### Two products, one control layer

**Vishwa CLI** is an agent-native console for on-chain finance across Solana, Base, and Pharos. You describe intent in plain English - check a wallet, browse Polymarket prediction markets, place or close a trade - and every action clears your sealed constraint policy before it executes. See the Vishwa CLI quickstart.

**Naro BTCvp Vault** is a guardrailed BTC yield vault on Pharos. Depositors mint BTCvp, a non-rebasing token backed by native BTC, while an AI coordinator (Naro) manages allocation within a curated strategy framework (Ciara) and an independent verification layer (Veta) checks every capital movement before execution. See the Vault Overview page.

### Who it's for

* **Individuals running agents.** Let an agent trade or manage funds inside hard limits you set once - per-trade caps, venue allow-lists, slippage bounds.
* **BTC holders.** Earn yield on native BTC without wrapping or bridging it, with every capital movement verified against predefined constraints.
* **Agent and strategy builders.** Build on a gateway where your agent proposes and independent infrastructure decides, so a bug or a hallucination cannot move funds outside policy.
* **Institutions.** Deploy capital through AI-coordinated strategies with enforced risk limits, auditable execution, and no key handover.

### Why a control layer, not just a smarter agent

Wallet policy engines see the shape of a transaction but not the market. Vishwa's control layer verifies intent, solvency, and execution fidelity - it pulls oracle and venue prices itself, so it can catch a bad fill that a transaction-level check cannot see. The verification proof replaces the human approval loop: no after-the-fact alerts, no irreversible surprises.

Nothing in this documentation is financial advice. Prediction markets and yield strategies carry risk.