> For the complete documentation index, see [llms.txt](https://docs.useshadowpay.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.useshadowpay.com/agents-and-automation/agent-accounts-overview.md). # Agent Accounts Overview ShadowPay treats AI agents as first-class account holders. An agent doesn't get bolted onto a human account through an API key and a webhook. It gets its own wallet, its own identity, and its own transaction history, governed by rules you define. *** ## Agent identity Every agent account has: * Its own Solana keypair, distinct from its parent account's keypair * A ShadowPay Agent ID, namespaced to its parent (`@yourname/agent-name`) * Its own encrypted balance under Confidential Balances * A transaction history that rolls up into the parent account's feed in real time An agent can hold USDC, receive funding from its parent, and transact autonomously within the limits it's been given. *** ## The spend lifecycle ``` Human creates Agent account ↓ Human defines a spend policy (limits, allowed recipients, time windows) ↓ Human funds the agent's wallet ↓ Agent evaluates a task and decides to transact ↓ Spend policy is checked before the agent's transaction is ever signed ↓ Within limits: agent signs and submits the transaction directly ↓ Above the human-in-the-loop threshold: push notification sent, human approves or rejects ↓ Transaction settles on Solana → webhook fires → parent feed updates ``` Spend policy enforcement happens before signing, not after. An agent physically cannot produce a valid transaction that violates its own policy limits. See [Spend Policies](/agents-and-automation/spend-policies.md) for the mechanics. *** ## Use cases * AI coding agents paying for LLM API calls as they run, instead of relying on a prepaid credit balance * Autonomous trading agents executing swaps within a defined risk budget * Agentic SaaS platforms billing usage per sub-task in near real time * Developer tools metering API consumption directly to the agents consuming it, without an intermediary billing system *** ## Privacy for agents Agent transactions are confidential by default, exactly like human transactions. This matters more than it might first appear: an agent's spend pattern, if public, can leak information about a business's usage volume, cost structure, or competitive activity. ShadowPay keeps agent transaction amounts encrypted while keeping the agent's public address and the fact that a transaction occurred fully visible on-chain, same as any other account. *** ## Next * [Spend Policies](/agents-and-automation/spend-policies.md) * [x402 Payments](/agents-and-automation/x402-payments.md) * [Webhooks & Events](/agents-and-automation/webhooks-and-events.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.useshadowpay.com/agents-and-automation/agent-accounts-overview.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.