# Agent Spending Without Controls — Rodrigo Coelho & Pranav Maheshwari, Edge & Node

## Executive summary

The video outlines the critical need for a dedicated infrastructure layer—specifically focusing on micropayments and compliance controls—to enable agentic commerce. While Large Language Models (LLMs) and agents can transact at machine speed, traditional payment rails were designed for human decision loops. The speaker argues that as AI agents become more powerful through paid Machine Capability Platforms (MCPs), the system must incorporate robust compliance layers to handle sanctions screening and prevent overspending, making governance a prerequisite for enterprise adoption.

## Key takeaways

- Agentic Payments Require Paid Infrastructure: Agents are becoming increasingly powerful by utilizing specialized, often paid, MCP servers. The speaker argues that relying on free services is unsustainable; agents require an aggregator/skill file (like Ampersend) to manage payments in the background, enabling transactions with a payment gateway.
- Traditional Rails Fail at Machine Speed: Existing financial systems assume a human 'in the loop' for decision-making. Agents transact around the clock at machine speed, necessitating new protocols that can handle high-volume, automated payments and compliance checks.
- Compliance is the Enterprise Bottleneck: For large enterprises to adopt agentic systems (handling trillions of dollars), a mandatory compliance layer must be implemented. This layer verifies counterparty identity, screens for sanctions, and prevents malicious transactions, which is currently lacking in automated agent workflows.

## Technical details

- Blockchain Data Indexing & Micropayments: Edge & Node built The Graph protocol (a blockchain data indexing protocol) and pioneered a query micropayment system in 2021, referencing the HTTP 402 spec years before its industry adoption.
- Payment Standards and Protocols: The speaker notes that the X42 specification (released by Coinbase) is relevant to agentic payments. The system must integrate paid MCP tools and a payment gateway/wallet infrastructure to facilitate transactions.
- Agent Capabilities & Tools: Agents are superpowered by connecting them to specialized MCPs (e.g., Exop, Firecrawl). The speaker demonstrates that an agent with a paid skill file can find specific email addresses and execute complex e-commerce transactions (like buying gifts) without manual input.
- Compliance Layer Functionality: The compliance layer must scan wallet addresses to ensure they are not sanctioned or involved in malicious activity. The demo showed that disabling screening allowed both good and bad (sanctioned) wallets to transact; enabling it blocked the bad wallet.

## Practical implications

- Build engineers must design payment pipelines that integrate real-time compliance and sanctions screening (e.g., using a dedicated middleware layer).
- Agent workflows need to be designed around paid API/MCP integrations rather than assuming free access.
- The architecture must support automated, high-speed transactions while maintaining auditable financial governance.

## Topics

AI Agents, Agentic Commerce, Micropayments, Blockchain Infrastructure, Financial Compliance, LLMs, The Graph Protocol, Ampersend, X42 spec, HTTP 402 spec

Source: https://www.youtube.com/watch?v=ZyGMqdIpPoE