aegis-financial-agent

Execute USDC transfers, swaps, CCTP bridging, and yield actions via the Aegis REST API on Arc Testnet.

1|Updated May 24, 2026
One-click install
npx skills add https://github.com/Wizbisy/aegis-all --skill aegis-financial-agent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aegis-financial-agent
Source: https://github.com/Wizbisy/aegis-all/tree/main/aegis/public
Command: npx skills add https://github.com/Wizbisy/aegis-all --skill aegis-financial-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of AI agents being financially paralyzed by unsafe, non-deterministic, and error-prone wallet interactions that can lead to double-spends, unauthorized transfers, or hallucinated transaction execution.

Core Features & Use Cases

  • Security-hardened intent execution: Executes financial intents via Aegis’ REST API on Arc Testnet with idempotency, nonces, rate limiting, and immutable audit logging.
  • Core financial operations: Supports USDC transfers, same-chain swaps, CCTP cross-chain bridging, x402 micropayments, and yield vault deposit/withdraw.
  • Policy and cost awareness: Guides the agent to check balances first, enforce spending policies, and estimate fees/slippage before any mutation, including async bridge polling until completion.

Real-world use case: An autonomous agent receives the intent “Move 50 USDC to Ethereum Sepolia and then swap any idle USDC to EURC,” and the skill enforces nonce/idempotency, confirms with the user before mutations, estimates costs, and polls the bridge audit until success or failure.

Quick Start

Use the aegis-financial-agent skill to transfer 25.00 USDC to a specified Arc Testnet EVM destination, by asking for balances, estimating costs, confirming with the user, then performing the transfer with proper nonce and idempotency headers.

Frequently Asked Questions about aegis-financial-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute autonomous cross-chain bridging for AI agents without risking double-spends?

Autonomous cross-chain bridging uses CCTP protocol routing with nonce and UUID v4 idempotency handshakes to prevent double-spends. The agent enforces balance checks, estimates fees, and mandates async polling until the bridge status reaches SUCCESS or FAILED.

What is idempotency in financial operations and why do AI agents need it for token swaps?

Idempotency in financial operations ensures mutations execute only once by combining a nonce with a UUID v4 identifier. AI agents need this handshake for token swaps to prevent duplicate or unauthorized transactions caused by network retries or hallucinated execution logic.

How to set up x402 micropayments for an autonomous agent on Arc Testnet?

Setting up x402 micropayments requires routing intents through the Aegis REST API on Arc Testnet with required authentication headers. The agent must check balances and enforce spending policies before executing the mutation with the nonce and idempotency handshake.

Does yield vault participation support automated deposit and withdrawal for AI agents?

Yield vault participation supports automated deposit and withdrawal for AI agents by routing wealth intents to the Aegis REST API. The agent enforces required authentication, applies idempotency handshakes for mutations, and checks balances before executing yield actions.

Can I place limit orders and DCA intents for USDC safely through an autonomous agent?

You can place limit orders and DCA intents for USDC safely by routing wealth intents to the Aegis REST API. The agent enforces spending policies, estimates slippage and fees, confirms with the user before mutations, and applies immutable audit logging.

When should I not use autonomous financial agents for onchain transactions?

You should not use autonomous financial agents for onchain transactions when required authentication headers, balance checks, or policy validations cannot be met. The agent mandates cost estimation and user confirmation before mutations, preventing execution in unsafe or non-deterministic environments.