# POINTS for Agents
> Connect over MCP, read policy, and get a signed authorization. Approve does not execute. Execution: not enabled.

## Docs
- [For Agents](https://thepoints.io/docs/agents.md): Connect an AI agent to POINTS over MCP, read operator policy, and get a human-approved signed authorization. Approve does not execute.
- [Concepts](https://thepoints.io/docs/agents/concepts.md): How Agent Pay authorization works: envelopes, lifecycle, scoped approval, recurring mandates, cash buckets, and payment patterns.
- [Authorization envelope](https://thepoints.io/docs/agents/concepts/authorization-envelope.md): The signed object POINTS records when a human approves a payment draft, and what it binds.
- [Cash buckets](https://thepoints.io/docs/agents/concepts/cash-buckets.md): How POINTS splits a balance into unreserved, reserved, and escrowed cash so pending drafts are never counted as spendable.
- [Lifecycle](https://thepoints.io/docs/agents/concepts/lifecycle.md): Approval windows, executableUntil, read-time expiry, revocation, the revoked status, and the zero clock-skew rule.
- [Payment patterns](https://thepoints.io/docs/agents/concepts/payment-patterns.md): Escrow-before-work, charge-on-2xx with fee-before-send, and graduated autonomy. All require execution (not enabled).
- [Recurring mandates](https://thepoints.io/docs/agents/concepts/recurring-mandates.md): Recurring mandates. Every cycle needs a fresh human approval, early claims are refused, and a claim never moves money.
- [Scoped approval](https://thepoints.io/docs/agents/concepts/scoped-approval.md): The spend envelope, the approval ledger, and approval scope, with today's limits.
- [Discovery](https://thepoints.io/docs/agents/discovery.md): How an agent finds POINTS. MCP endpoint, OAuth metadata, llms files, status.json, catalog.json, and the reference JSON.
- [FAQ](https://thepoints.io/docs/agents/faq.md): Straight answers on replay, price drift, self-approval, receipts, and disputes for agents using POINTS.
- [Hosts](https://thepoints.io/docs/agents/hosts.md): Which MCP hosts work with POINTS today and which are planned next. ChatGPT is live; Claude, Cursor and OpenClaw-class runtimes are planned.
- [ChatGPT](https://thepoints.io/docs/agents/hosts/chatgpt.md): Connect ChatGPT to the Points MCP server in Developer Mode, approve scopes, and check the connection.
- [Claude](https://thepoints.io/docs/agents/hosts/claude.md): Claude Desktop and Claude Code support for the POINTS MCP server is planned. Use ChatGPT today, or follow the discovery guide for any MCP host.
- [Cursor](https://thepoints.io/docs/agents/hosts/cursor.md): Cursor support for the POINTS MCP server is planned. Use ChatGPT today, or follow the discovery guide for any MCP host.
- [OpenClaw](https://thepoints.io/docs/agents/hosts/openclaw.md): Support for OpenClaw-class agent runtimes on the POINTS MCP server is planned. Use ChatGPT today, or follow the discovery guide.
- [Quickstart](https://thepoints.io/docs/agents/quickstart.md): Connect over MCP with OAuth, read the policy vocabulary, prepare a payment, get human approval, and check its status. Nothing executes.
- [MCP & API reference](https://thepoints.io/docs/agents/reference.md): Tools, scopes, statuses and policy vocabulary as JSON.
- [Errors](https://thepoints.io/docs/agents/reference/errors.md): prepare_payment failure classes, JSON-RPC errors, and tool errors. Generated from code.
- [MCP tools](https://thepoints.io/docs/agents/reference/mcp-tools.md): Every Points MCP tool with its OAuth scope, annotations, and input schema. Names, scopes, and schemas generated from code.
- [Policy vocabulary](https://thepoints.io/docs/agents/reference/policy-vocabulary.md): Policy schema version, canonical categories, matching rules, and default velocity caps. Generated from code.
- [OAuth scopes](https://thepoints.io/docs/agents/reference/scopes.md): The OAuth scopes the Points MCP server supports and which tool needs which. Generated from code.
- [Statuses](https://thepoints.io/docs/agents/reference/statuses.md): Payment draft, approval, envelope, mandate, and wallet status values with their meaning. Generated from code.
- [Wallet status](https://thepoints.io/docs/agents/wallet-status.md): Read an AI Agent's Crossmint staging wallet status with get_wallet_status. Status only, with no balances or money movement.

## Machine-readable
- [Agent Pay status](https://thepoints.io/docs/agents/status.json): Execution and settlement state, derived at request time.
- [MCP tools](https://thepoints.io/docs/agents/reference/mcp-tools.json): Tools with scopes, annotations, and input schemas, generated from code.
- [OAuth scopes](https://thepoints.io/docs/agents/reference/scopes.json): Scopes and the tools that need them.
- [Statuses](https://thepoints.io/docs/agents/reference/statuses.json): Payment draft, approval, envelope, mandate, and wallet status values.
- [Policy vocabulary](https://thepoints.io/docs/agents/reference/policy-vocabulary.json): Policy schema version, categories, matching rules, and default caps.
- [Errors](https://thepoints.io/docs/agents/reference/errors.json): prepare_payment failure classes and protocol errors.
- [Product catalog](https://thepoints.io/docs/catalog.json): Every POINTS product with its status.
- [Changelog](https://thepoints.io/changelog.json): What shipped to production, by commit.
- [MCP endpoint](https://thepoints.io/mcp): Streamable HTTP; discover OAuth metadata first.
- [OAuth protected resource metadata](https://thepoints.io/.well-known/oauth-protected-resource)
- [OAuth authorization server metadata](https://thepoints.io/.well-known/oauth-authorization-server)

## Optional
- [All docs as one file](https://thepoints.io/llms-full.txt)