# Conto

Conto is a buyer-side control layer between an agent and transaction execution. The agent owner
or budget holder sets spending limits, permitted recipients, and approval rules. The buyer runtime
checks Conto before paying; merchants do not install Conto. Signing and credential controls depend
on the connected execution path. A recipient request never grants authority to spend.

## Read first

- [Buyer-side control](https://conto.finance/docs/guides/buyer-side-control): owner authority, buyer integration, and execution boundaries.
- [Quickstart](https://conto.finance/docs/quickstart/setup): create an agent with the CLI, send a Tempo Testnet payment, and verify settlement.
- [Agent sandbox](https://conto.finance/docs/quickstart/agent-sandbox): no-account test path for autonomous agents and CI-style evaluation.
- [Choose your integration](https://conto.finance/docs/guides/choose-your-integration): SDK, REST, MCP, skills, x402, MPP, and wallet-model routing.

## Integration surfaces

- [SDK install](https://conto.finance/docs/sdk/installation): TypeScript client setup.
- [Authentication](https://conto.finance/docs/sdk/authentication): organization API keys vs agent SDK keys.
- [Payments](https://conto.finance/docs/sdk/payments): request, execute, approve, confirm, status, and autoExecute.
- [Admin SDK](https://conto.finance/docs/sdk/admin): provision agents, wallets, policies, and SDK keys.
- [CLI](https://conto.finance/docs/cli/quickstart): guided and headless setup.
- [MCP](https://conto.finance/docs/mcp/overview): tool server for assistant/operator workflows.
- [Agent skills](https://conto.finance/docs/sdk/skills): OpenClaw and Hermes policy enforcement.

## Common tasks

- [Testing policies](https://conto.finance/docs/guides/policy-testing): prove approve, review, and deny outcomes.
- [x402 API payments](https://conto.finance/docs/guides/x402-api-payments): pay-per-request API flow.
- [MPP sessions](https://conto.finance/docs/guides/mpp-session-payments): session-based micropayments.
- [Conto Pay quickstart](https://conto.finance/docs/conto-pay/quickstart): open the hosted workspace and run a first test payment.
- [Track and recover a payment](https://conto.finance/docs/guides/payment-lifecycle): reconcile settlement, human review, required actions, and uncertain execution.
- [Recipes](https://conto.finance/docs/guides/recipes): copy-paste curl snippets.

## Reference

- [API Reference](https://conto.finance/docs/api/reference): OpenAPI, Swagger, base URLs, and credential routing.
- [OpenAPI JSON](https://conto.finance/api/openapi): machine-readable schema.
- [API conventions](https://conto.finance/docs/reference/api-conventions): amount units, IDs, pagination, and endpoint-specific retries.
- [Credential scopes](https://conto.finance/docs/reference/auth-scopes): generated standard and admin agent-key presets.
- [Payment flows](https://conto.finance/docs/reference/payment-support): signing, network, and environment prerequisites.
- [Defaults](https://conto.finance/docs/reference/defaults): API, SDK, and CLI defaults.
- [Full docs text](https://conto.finance/llms-full.txt): complete docs, blog content, and generated API summaries.

## Rules for agents

- Use an agent SDK key (`conto_agent_...`) for `/api/sdk/*` payment and agent-scoped reads.
- Use an organization API key (`conto_...`) for provisioning agents, wallets, policies, and SDK keys.
- Use the CLI sandbox path when no human account or browser session is available.
- Do not execute payments that Conto returns as `DENIED` or `REQUIRES_APPROVAL`.

## Agent-Readable Site

- [Terminal site](https://conto.finance/ai): Markdown-style terminal view of the public Conto site.
- [Raw Markdown index](https://conto.finance/.md): Machine-readable index of public Conto pages.
- [Agent setup prompt](https://conto.finance/agents.md): Operational instructions for agents to discover, install, verify, and use Conto.

- [Conto](https://conto.finance/.md): Payment controls for agents before they spend, with wallets, policies, approvals, trust, and auditability for agentic payments.
- [Build With Conto](https://conto.finance/build.md): The agent and developer hub for SDK, MCP, CLI, OpenAPI, skills, llms.txt, and sandbox setup.
- [Conto Product](https://conto.finance/product.md): The Conto product surface for runtime policy, agent wallets, approvals, trust, audit, and developer integrations.
- [Conto Pay](https://conto.finance/product/conto-pay.md): Hosted payment agent workspace with profiles, checkout links, approvals, hosted requests, receipts, and assistant-led operations.
- [Conto Solutions](https://conto.finance/solutions.md): Use cases for autonomous payment operations, runtime governance, approvals, counterparty controls, and machine-speed budgets.
- [Conto Pricing](https://conto.finance/pricing.md): Pricing and plan information for Conto agentic payment controls.
- [Conto API Docs](https://conto.finance/api-docs.md): Interactive API documentation backed by the Conto OpenAPI schema.
- [Conto Security](https://conto.finance/security.md): Security disclosure, safe harbor, and machine-readable security information for Conto.
- [Conto for OpenClaw](https://conto.finance/openclaw.md): Installable Conto policy enforcement for OpenClaw agents.
- [Conto for Hermes](https://conto.finance/hermes.md): Installable Conto policy enforcement for Nous Hermes agents.
- [Try Conto](https://conto.finance/try.md): Public sandbox experience for trying Conto agent payment workflows.
- [Conto Demos](https://conto.finance/demos.md): Interactive demos for agentic payment workflows across industries and use cases.
- [Conto Resources](https://conto.finance/resources.md): Downloadable and browsable resources for Conto products, security, and agentic payment education.
- [Learn Agentic Payments](https://conto.finance/learn.md): Educational guides for agentic payments, x402, MPP, stablecoins, and policy controls.
- [Conto Partners](https://conto.finance/partners.md): Wallet, settlement, and trust partners helping teams operate controlled agent payments.
- [Contact Conto](https://conto.finance/contact.md): Contact Conto for sales, support, integration guidance, and production deployment conversations.
- [Policy Engine](https://conto.finance/product/policy-engine.md): As the agent owner or budget holder, decide how much it can spend, who it can pay, and when it must ask. Conto checks your policy before execution.
- [Agent Wallets](https://conto.finance/product/agent-wallets.md): Manage agent balances, track payments, and freeze wallets when spending needs to stop.
- [Approvals](https://conto.finance/product/approvals.md): Send payments that need approval to a reviewer before money moves.
- [Network Intelligence](https://conto.finance/product/network-intelligence.md): Use transaction history, relationship data, and shared risk signals to understand who agents are paying before the payment clears.
- [Audit and Reconciliation](https://conto.finance/product/audit-reconciliation.md): Give finance a record of each agent payment, from the initial request to the final result.
- [Developer Platform](https://conto.finance/product/developer-platform.md): Connect your agent to Conto and test its first payment in sandbox.
- [Policy-checked payouts](https://conto.finance/solutions/instant-payouts.md): Pay people or move funds as soon as work completes, with every payout checked before release.
- [Payment controls for autonomous procurement](https://conto.finance/solutions/autonomous-procurement.md): Agents select suppliers. Conto checks payments against your recipient rules, spending limits, and approval requirements.
- [Refunds and credits within your rules](https://conto.finance/solutions/refunds-and-credits.md): Cap what an agent can credit per case and per day. Above it, a person decides.
- [Approvals for payment exceptions](https://conto.finance/solutions/approval-routing.md): Set the line above which a payment waits for a person, and where that request lands.
- [Vendor and counterparty controls](https://conto.finance/solutions/vendor-and-counterparty-controls.md): Pay approved vendors automatically. Review payments to new, changed, or risky recipients.
- [Budgets for machine payments](https://conto.finance/solutions/micropayments.md): Set spending limits per request and per session for API, inference, and compute usage.
- [Spend controls for personal assistants](https://conto.finance/solutions/personal-assistants.md): Each user sets which merchants their assistant can use, how much, and when to ask.
- [What Are Agentic Payments?](https://conto.finance/learn/what-are-agentic-payments.md): How AI agents buy, book, refund, pay APIs, and trigger payouts under policy.
- [Spend Management for AI Agents](https://conto.finance/learn/spend-management-for-ai-agents.md): Budgets, recipient rules, and approval thresholds for agents that pay for work, and where each check has to run.
- [How to Control AI Agent Spending](https://conto.finance/learn/how-to-control-ai-agent-spending.md): Budgets, policies, approvals, wallets, counterparties, and audit trails for agents that spend money.
- [x402 and API Spend Controls for AI Agents](https://conto.finance/learn/x402-api-spend-controls.md): How to keep paid API calls, data access, and machine-speed service usage inside policy before an agent pays.