Skip to content

How PayPal Agentic Commerce’s scores are calculated

The full audit trail, recomputed from the verdict data at build time through the same code that produced the leaderboard: verdict × quality × story weight per cell, cells sum to dimension scores, dimensions blend into the PA Score. Every number on the product page is reproducible from this page alone; for why the formula looks like this, see the methodology.

verdict factors: full ×1.0 · partial ×0.6 · disputed ×0.3 · none ×0.0 · n/a excluded from both sides · cell points = weight × quality × factor · cell max = weight × 10

PA Score27/100

Agent-ready 61.7 × 0.30 = 18.51

API quality 11.4 × 0.20 = 2.28

Openness 18.0 × 0.20 = 3.60

Built-in AI 16.0 × 0.15 = 2.40

Automation 0.0 × 0.15 = 0.00

(18.51 + 2.28 + 3.60 + 2.40 + 0.00) ÷ (0.30 + 0.20 + 0.20 + 0.15 + 0.15) = 26.79 ÷ 1.00 = 26.8

Scores are stored to 1 decimal; the product page’s pills round to whole numbers for display. Each dimension below shows the stories, verdicts, and cited evidence behind its number.

Agent-ready61.7/100×0.30 of the PA blend

Outside-in: can YOUR agent reach and drive this product — API, MCP, CLI, headless runs, agent docs.

Point an agent at llms.txt or agent-oriented docsweight 2

2 (weight) × 9 (quality) × 1.0 (full) = 18.0 of 20 max

  • [probe] https://developer.paypal.com/llms.txtPROBE llms.txt: HTTP 200 at https://developer.paypal.com/llms.txt # PayPal Developer Documentation > PayPal's developer documentation for integrating payments, managing transactions, an
  • [probe] https://developer.paypal.com/agent-ready/overview.mdPROBE docs-md: HTTP 200 at https://developer.paypal.com/agent-ready/overview.md # Agent Ready overview (/agent-ready/overview) Agent Ready helps Braintree merchants accept payments from AI shopping
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverofficial MCP server documented at https://developer.paypal.com/ai-tools/mcp-server
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).

Run the product headlessly / in CI for automationweight 2

2 (weight) × 6 (quality) × 0.6 (partial) = 7.2 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [probe] https://developer.paypal.com/ai-tools/toolkitPROBE runtime (recorded 2026-09-14): the PayPal Agent Toolkit resolves on the public npm registry (npm view @paypal/agent-toolkit → 1.11.0, published 2026-09-01; Python twin paypal-agent-toolkit 1.11.0 on PyPI) — the SDK covering orders, invoices, subscriptions, tracking, disputes, refunds, and transactions for MCP, OpenAI Agents SDK, LangChain, CrewAI, Vercel AI SDK, and Bedrock.
  • [claimed-docs] https://developer.paypal.com/tools/sandbox/The PayPal sandbox "is a self-contained, virtual testing environment that simulates the live PayPal production environment" — "a shielded space where you can initiate and watch while your apps process PayPal API requests without touching any live PayPal acc[ounts]"; "the sandbox mirrors the features on the PayPal production servers", with sandbox accounts and "their associated authentication credentials" used "in your PayPal API calls" so "you can test and debug your apps" before going live. Agent-commerce flows (cart API, agent toolkit tools) run against the same REST surface.
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.

Plug MCP servers into this product so it can use their toolsweight 3

n/a — not applicable to this product: excluded from numerator and denominator

no evidence cited — the verdict rests on absence of evidence, re-checked on refresh

Connect an agent via an official MCP serverweight 3

3 (weight) × 8 (quality) × 1.0 (full) = 24.0 of 30 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/mcp-serverPayPal built an MCP server that lets merchants use natural language with their favorite MCP clients. This helps users complete business tasks more easily.
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverofficial MCP server documented at https://developer.paypal.com/ai-tools/mcp-server
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).
  • [probe] https://developer.paypal.com/ai-tools/toolkitPROBE runtime (recorded 2026-09-14): the PayPal Agent Toolkit resolves on the public npm registry (npm view @paypal/agent-toolkit → 1.11.0, published 2026-09-01; Python twin paypal-agent-toolkit 1.11.0 on PyPI) — the SDK covering orders, invoices, subscriptions, tracking, disputes, refunds, and transactions for MCP, OpenAI Agents SDK, LangChain, CrewAI, Vercel AI SDK, and Bedrock.
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."

Use an official CLIweight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

no evidence cited — the verdict rests on absence of evidence, re-checked on refresh

Drive the product through a documented public APIweight 3

3 (weight) × 9 (quality) × 1.0 (full) = 27.0 of 30 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkitPayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using Amazon Bedrock, CrewAI, LangChain, Model Context Protocol (MCP), OpenAI's Agents SDK, and Vercel's AI SDK.
  • [claimed-docs] https://developer.paypal.com/ai-tools/mcp-serverPayPal built an MCP server that lets merchants use natural language with their favorite MCP clients. This helps users complete business tasks more easily.
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/createcartCreates a new cart with the provided PayPalCart object. This is called by PayPal Shopping Cart when an AI agent or customer initiates a checkout process.
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/getcartRetrieves the current state of a cart. Used by PayPal Shopping Cart to check cart status, validation issues, and current totals.
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).
  • [probe] https://developer.paypal.com/ai-tools/toolkitPROBE runtime (recorded 2026-09-14): the PayPal Agent Toolkit resolves on the public npm registry (npm view @paypal/agent-toolkit → 1.11.0, published 2026-09-01; Python twin paypal-agent-toolkit 1.11.0 on PyPI) — the SDK covering orders, invoices, subscriptions, tracking, disputes, refunds, and transactions for MCP, OpenAI Agents SDK, LangChain, CrewAI, Vercel AI SDK, and Bedrock.
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.
  • [claimed-docs] https://developer.paypal.com/tools/sandbox/The PayPal sandbox "is a self-contained, virtual testing environment that simulates the live PayPal production environment" — "a shielded space where you can initiate and watch while your apps process PayPal API requests without touching any live PayPal acc[ounts]"; "the sandbox mirrors the features on the PayPal production servers", with sandbox accounts and "their associated authentication credentials" used "in your PayPal API calls" so "you can test and debug your apps" before going live. Agent-commerce flows (cart API, agent toolkit tools) run against the same REST surface.

Issue scoped/least-privilege API credentials for an agentweight 2

2 (weight) × 4 (quality) × 0.6 (partial) = 4.8 of 20 max

  • [claimed-docs] https://developer.paypal.com/agent-ready/agentic-commerce-protocolThe ACP path is an in-ChatGPT checkout build: "Implement an MCP server with the `complete_checkout` tool to receive tokens", "Call `requestCheckout()` from the app to trigger Instant Checkout", "Follow the ACP agentic checkout specification to manage checkout sessions", and "Specify `braintree` as your payment provider." The delegated credential "is a one-time-use token that you can process using your existing Braintree integration" — a Braintree payment method nonce that "serves as a secure, single-use reference to the buyer's payment information" and is "bound to your merchant ID and includes amount and time restrictions that you can configure."
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/createcartCart execution is a reverse API the merchant hosts: POST /merchant-cart "Creates a new cart with the provided PayPalCart object. This is called by PayPal Shopping Cart when an AI agent or customer initiates a checkout process", returning "201 Created: Cart created successfully with payment token in payment_method" or 200 plus validation_issues ("Business logic issues that need resolution"). Authentication is PayPal-issued, not merchant-issued: "These JWT tokens are issued and managed by PayPal, not generated by merchants", and the merchant must "Verify token signature using PayPal's public keys".

Build against official SDKsweight 2

2 (weight) × 8 (quality) × 1.0 (full) = 16.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkitPayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using Amazon Bedrock, CrewAI, LangChain, Model Context Protocol (MCP), OpenAI's Agents SDK, and Vercel's AI SDK.
  • [claimed-docs] https://developer.paypal.com/ai-tools/mcp-serverPayPal built an MCP server that lets merchants use natural language with their favorite MCP clients. This helps users complete business tasks more easily.
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkitIntegrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions.
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [claimed-docs] https://developer.paypal.com/ai-tools/agent-toolsPost-purchase operations are first-class agent tools: refunds via create_refund / get_refund (a "refund for a captured payment", where "If not specified, a refund is for the full amount of the purchase"), disputes via list_disputes ("Retrieve a summary of all disputes with optional filtering"), get_dispute, and accept_dispute_claim ("Accept a dispute claim, resolving it in favor of the buyer."), and orders via create_order ("Create an order in the PayPal system based on the provided details."), pay_order, and create_shipment. Catalog-management tools (create_product, list_product, show_product_details) let agents write to the PayPal product catalog itself.
  • [probe] https://developer.paypal.com/ai-tools/toolkitPROBE runtime (recorded 2026-09-14): the PayPal Agent Toolkit resolves on the public npm registry (npm view @paypal/agent-toolkit → 1.11.0, published 2026-09-01; Python twin paypal-agent-toolkit 1.11.0 on PyPI) — the SDK covering orders, invoices, subscriptions, tracking, disputes, refunds, and transactions for MCP, OpenAI Agents SDK, LangChain, CrewAI, Vercel AI SDK, and Bedrock.
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).
  • [claimed-docs] https://developer.paypal.com/tools/sandbox/The PayPal sandbox "is a self-contained, virtual testing environment that simulates the live PayPal production environment" — "a shielded space where you can initiate and watch while your apps process PayPal API requests without touching any live PayPal acc[ounts]"; "the sandbox mirrors the features on the PayPal production servers", with sandbox accounts and "their associated authentication credentials" used "in your PayPal API calls" so "you can test and debug your apps" before going live. Agent-commerce flows (cart API, agent toolkit tools) run against the same REST surface.
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.
  • [github] https://github.com/paypal/agent-toolkitcreate_order: Create an order in PayPal system based on provided details

Subscribe to events via webhooksweight 2

2 (weight) × 7 (quality) × 1.0 (full) = 14.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/api/webhooks/overview"Webhooks are HTTPS posts from PayPal to an endpoint on your server whenever the corresponding event type occurs" — "Up to 10 webhook URLs may be subscribed per app", each scoped to specific event types or all of them. Delivery is durable: "Any non-2xx status code will cause PayPal to reattempt delivery up to 25 times over the course of 3 days or until it receives a 2xx success code." Authenticity checks are documented ("Post the message, stored webhook ID, and header information back to PayPal's verify signature endpoint"), and the warning is explicit: "If you do not verify a message, you will not have a way to validate that its sender was in fact PayPal."
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.

Agent-ready = 111.0 ÷ 180 × 100 = 61.7

API quality11.4/100×0.20 of the PA blend

The programmable surface once an agent is there — machine-readable spec, interactive docs, sandbox, versioning discipline.

Explore an interactive API reference with runnable examplesweight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/createcartCreates a new cart with the provided PayPalCart object. This is called by PayPal Shopping Cart when an AI agent or customer initiates a checkout process.
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/getcartRetrieves the current state of a cart. Used by PayPal Shopping Cart to check cart status, validation issues, and current totals.
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.
  • [claimed-docs] https://developer.paypal.com/tools/sandbox/The PayPal sandbox "is a self-contained, virtual testing environment that simulates the live PayPal production environment" — "a shielded space where you can initiate and watch while your apps process PayPal API requests without touching any live PayPal acc[ounts]"; "the sandbox mirrors the features on the PayPal production servers", with sandbox accounts and "their associated authentication credentials" used "in your PayPal API calls" so "you can test and debug your apps" before going live. Agent-commerce flows (cart API, agent toolkit tools) run against the same REST surface.

Download a machine-readable API spec (OpenAPI or equivalent)weight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/createcartCart execution is a reverse API the merchant hosts: POST /merchant-cart "Creates a new cart with the provided PayPalCart object. This is called by PayPal Shopping Cart when an AI agent or customer initiates a checkout process", returning "201 Created: Cart created successfully with payment token in payment_method" or 200 plus validation_issues ("Business logic issues that need resolution"). Authentication is PayPal-issued, not merchant-issued: "These JWT tokens are issued and managed by PayPal, not generated by merchants", and the merchant must "Verify token signature using PayPal's public keys".
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/getcartThe companion read endpoint keeps agent-side cart state fresh: get cart "Retrieves the current state of a cart. Used by PayPal Shopping Cart to check cart status, validation issues, and current totals." Together with create cart, this is the crawled surface of PayPal's agentic-commerce v1 cart API — PayPal's Shopping Cart service, not the AI agent directly, calls the merchant's endpoints.

Test against a sandbox environment without touching production dataweight 1

1 (weight) × 8 (quality) × 1.0 (full) = 8.0 of 10 max

  • [claimed-docs] https://developer.paypal.com/tools/sandbox/The PayPal sandbox "is a self-contained, virtual testing environment that simulates the live PayPal production environment" — "a shielded space where you can initiate and watch while your apps process PayPal API requests without touching any live PayPal acc[ounts]"; "the sandbox mirrors the features on the PayPal production servers", with sandbox accounts and "their associated authentication credentials" used "in your PayPal API calls" so "you can test and debug your apps" before going live. Agent-commerce flows (cart API, agent toolkit tools) run against the same REST surface.
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/createcartCart execution is a reverse API the merchant hosts: POST /merchant-cart "Creates a new cart with the provided PayPalCart object. This is called by PayPal Shopping Cart when an AI agent or customer initiates a checkout process", returning "201 Created: Cart created successfully with payment token in payment_method" or 200 plus validation_issues ("Business logic issues that need resolution"). Authentication is PayPal-issued, not merchant-issued: "These JWT tokens are issued and managed by PayPal, not generated by merchants", and the merchant must "Verify token signature using PayPal's public keys".
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."

Rely on versioned APIs with a documented deprecation policyweight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/createcartCreates a new cart with the provided PayPalCart object. This is called by PayPal Shopping Cart when an AI agent or customer initiates a checkout process.
  • [claimed-docs] https://developer.paypal.com/api/agentic-commerce/v1/getcartRetrieves the current state of a cart. Used by PayPal Shopping Cart to check cart status, validation issues, and current totals.
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.

API quality = 8.0 ÷ 70 × 100 = 11.4

Openness18.0/100×0.20 of the PA blend

Can you leave, inspect, or self-host — data export, open source, portability.

Do everything through the API that I can do in the UIweight 2

2 (weight) × 6 (quality) × 0.6 (partial) = 7.2 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [claimed-docs] https://developer.paypal.com/ai-tools/agent-toolsPost-purchase operations are first-class agent tools: refunds via create_refund / get_refund (a "refund for a captured payment", where "If not specified, a refund is for the full amount of the purchase"), disputes via list_disputes ("Retrieve a summary of all disputes with optional filtering"), get_dispute, and accept_dispute_claim ("Accept a dispute claim, resolving it in favor of the buyer."), and orders via create_order ("Create an order in the PayPal system based on the provided details."), pay_order, and create_shipment. Catalog-management tools (create_product, list_product, show_product_details) let agents write to the PayPal product catalog itself.
  • [claimed-docs] https://developer.paypal.com/agentic-commerce-services/aboutAccess to PayPal's agentic commerce services is form-gated, not self-serve: merchants must complete a form "to contact the AI team at PayPal and request access. The PayPal AI team will follow up after your form submission to guide you through onboarding." The pitch is "Connect once to reach many platforms", with catalog on-ramps through partners ("Connect your product listings to PayPal's partners like Wix, Cymbio, Commerce (BigCommerce & Feedonomics), and Shopware") and a merchant-control promise: with Store Sync "you stay in charge of your business. You control your brand appearance and customer communications for all AI-powered transactions".
  • [claimed-docs] https://developer.paypal.com/store-sync/overviewStore Sync is the full-catalog path, "enabling AI agents to discover your products, create and manage shopping carts, and complete purchases on behalf of customers through conversational interfaces", while "Your existing order management system receives the order just as it would from any other channel." Scope is narrow today: "Store Sync currently supports merchants who are selling physical goods to US-based customers in USD. It does not support digital goods, subscription products, or sales to customers who are outside the US at this time." Eligibility requires an existing PayPal merchant account with an Orders v2 integration or Braintree "with a compatible setup", plus the same form-gated access approval.
  • [claimed-docs] https://developer.paypal.com/agent-ready/universal-commerce-protocolThe UCP path targets Google surfaces: "This enables AI-powered checkout experiences through Google AI Mode (Gemini), where Google Pay returns a tokenized credential that Braintree can process, and your server uses it to process the payment." Tokenization moves off the client — "With UCP, Google shifts this to the server side" and "Google includes it in checkout-complete payload" — so when Google calls the merchant's /checkout-session/{id}/complete endpoint, "the payment payload includes the Braintree single-use token in the `credential.token` field". Enablement is Google-gated: "UCP merchant registration with Google. Contact Google for enablement."
  • [claimed-docs] https://developer.paypal.com/api/rest/The developer portal's REST API getting-started page anchors the agent surface: "REST APIs, including orders, payments, subscriptions, invoicing, and disputes", authenticated with OAuth 2 access tokens, alongside Server SDKs, API response Codes, and Webhook events references — the same rails the Agent Toolkit's tools call.
  • [claimed-docs] https://developer.paypal.com/tools/sandbox/The PayPal sandbox "is a self-contained, virtual testing environment that simulates the live PayPal production environment" — "a shielded space where you can initiate and watch while your apps process PayPal API requests without touching any live PayPal acc[ounts]"; "the sandbox mirrors the features on the PayPal production servers", with sandbox accounts and "their associated authentication credentials" used "in your PayPal API calls" so "you can test and debug your apps" before going live. Agent-commerce flows (cart API, agent toolkit tools) run against the same REST surface.

Export all of my data in open formats and leaveweight 3

n/a — not applicable to this product: excluded from numerator and denominator

no evidence cited — the verdict rests on absence of evidence, re-checked on refresh

Read the product's source under an open licenseweight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

no evidence cited — the verdict rests on absence of evidence, re-checked on refresh

Self-host the core productweight 3

n/a — not applicable to this product: excluded from numerator and denominator

no evidence cited — the verdict rests on absence of evidence, re-checked on refresh

Openness = 7.2 ÷ 40 × 100 = 18.0

Built-in AI16.0/100×0.15 of the PA blend

Inside-out: how agentic the product itself is for its users — built-in assistants, autonomous features.

Get AI-generated insights and suggestions from my data inside the productweight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [claimed-docs] https://developer.paypal.com/ai-tools/agent-toolsPost-purchase operations are first-class agent tools: refunds via create_refund / get_refund (a "refund for a captured payment", where "If not specified, a refund is for the full amount of the purchase"), disputes via list_disputes ("Retrieve a summary of all disputes with optional filtering"), get_dispute, and accept_dispute_claim ("Accept a dispute claim, resolving it in favor of the buyer."), and orders via create_order ("Create an order in the PayPal system based on the provided details."), pay_order, and create_shipment. Catalog-management tools (create_product, list_product, show_product_details) let agents write to the PayPal product catalog itself.
  • [claimed-docs] https://developer.paypal.com/ai-tools/mcp-serverPayPal built an MCP server that lets merchants use natural language with their favorite MCP clients. This helps users complete business tasks more easily.

Set up automations that run autonomously in the backgroundweight 2

2 (weight) × 6 (quality) × 0.6 (partial) = 7.2 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [claimed-docs] https://developer.paypal.com/api/webhooks/overview"Webhooks are HTTPS posts from PayPal to an endpoint on your server whenever the corresponding event type occurs" — "Up to 10 webhook URLs may be subscribed per app", each scoped to specific event types or all of them. Delivery is durable: "Any non-2xx status code will cause PayPal to reattempt delivery up to 25 times over the course of 3 days or until it receives a 2xx success code." Authenticity checks are documented ("Post the message, stored webhook ID, and header information back to PayPal's verify signature endpoint"), and the warning is explicit: "If you do not verify a message, you will not have a way to validate that its sender was in fact PayPal."
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).
  • [probe] https://developer.paypal.com/ai-tools/toolkitPROBE runtime (recorded 2026-09-14): the PayPal Agent Toolkit resolves on the public npm registry (npm view @paypal/agent-toolkit → 1.11.0, published 2026-09-01; Python twin paypal-agent-toolkit 1.11.0 on PyPI) — the SDK covering orders, invoices, subscriptions, tracking, disputes, refunds, and transactions for MCP, OpenAI Agents SDK, LangChain, CrewAI, Vercel AI SDK, and Bedrock.
  • [claimed-docs] https://developer.paypal.com/ai-tools/agent-toolsPost-purchase operations are first-class agent tools: refunds via create_refund / get_refund (a "refund for a captured payment", where "If not specified, a refund is for the full amount of the purchase"), disputes via list_disputes ("Retrieve a summary of all disputes with optional filtering"), get_dispute, and accept_dispute_claim ("Accept a dispute claim, resolving it in favor of the buyer."), and orders via create_order ("Create an order in the PayPal system based on the provided details."), pay_order, and create_shipment. Catalog-management tools (create_product, list_product, show_product_details) let agents write to the PayPal product catalog itself.
  • [claimed-docs] https://developer.paypal.com/agentic-commerce-services/aboutAccess to PayPal's agentic commerce services is form-gated, not self-serve: merchants must complete a form "to contact the AI team at PayPal and request access. The PayPal AI team will follow up after your form submission to guide you through onboarding." The pitch is "Connect once to reach many platforms", with catalog on-ramps through partners ("Connect your product listings to PayPal's partners like Wix, Cymbio, Commerce (BigCommerce & Feedonomics), and Shopware") and a merchant-control promise: with Store Sync "you stay in charge of your business. You control your brand appearance and customer communications for all AI-powered transactions".

Delegate tasks to a built-in AI assistant inside the productweight 3

3 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 30 max

no evidence cited — the verdict rests on absence of evidence, re-checked on refresh

Operate the product with natural-language commandsweight 2

2 (weight) × 6 (quality) × 0.6 (partial) = 7.2 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/mcp-serverPayPal built an MCP server that lets merchants use natural language with their favorite MCP clients. This helps users complete business tasks more easily.
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [claimed-docs] https://developer.paypal.com/ai-tools/agent-toolsPost-purchase operations are first-class agent tools: refunds via create_refund / get_refund (a "refund for a captured payment", where "If not specified, a refund is for the full amount of the purchase"), disputes via list_disputes ("Retrieve a summary of all disputes with optional filtering"), get_dispute, and accept_dispute_claim ("Accept a dispute claim, resolving it in favor of the buyer."), and orders via create_order ("Create an order in the PayPal system based on the provided details."), pay_order, and create_shipment. Catalog-management tools (create_product, list_product, show_product_details) let agents write to the PayPal product catalog itself.
  • [probe] https://developer.paypal.com/ai-tools/mcp-serverPROBE runtime (recorded 2026-09-14): PayPal's remote MCP server is live — a keyless JSON-RPC initialize POST to https://mcp.paypal.com/mcp answered HTTP 401 with 'www-authenticate: Bearer realm="OAuth", resource_metadata="https://mcp.paypal.com/.well-known/oauth-protected-resource/mcp"', and that RFC 9728 metadata URL answers keylessly, naming the resource and its authorization server (PKCE S256, dynamic client registration at /register).
  • [probe] https://developer.paypal.com/ai-tools/toolkitPROBE runtime (recorded 2026-09-14): the PayPal Agent Toolkit resolves on the public npm registry (npm view @paypal/agent-toolkit → 1.11.0, published 2026-09-01; Python twin paypal-agent-toolkit 1.11.0 on PyPI) — the SDK covering orders, invoices, subscriptions, tracking, disputes, refunds, and transactions for MCP, OpenAI Agents SDK, LangChain, CrewAI, Vercel AI SDK, and Bedrock.
  • [claimed-docs] https://developer.paypal.com/agent-ready/overview"Agent Ready helps Braintree merchants accept payments from AI shopping assistants across major platforms, including ChatGPT, Google AI Mode, and Gemini, without building separate integrations for each." Two protocol paths are supported: ACP for "ChatGPT, OpenAI" via "Delegated payment tokens through Braintree", and UCP for "Google AI Mode, Gemini" via the "Google Pay payment handler through Braintree". Both explicitly assume the merchant can "Leverage your existing Braintree merchant account and processing relationship" — the gate is Braintree, not a plain PayPal account.
  • [claimed-docs] https://developer.paypal.com/agentic-commerce-services/aboutAccess to PayPal's agentic commerce services is form-gated, not self-serve: merchants must complete a form "to contact the AI team at PayPal and request access. The PayPal AI team will follow up after your form submission to guide you through onboarding." The pitch is "Connect once to reach many platforms", with catalog on-ramps through partners ("Connect your product listings to PayPal's partners like Wix, Cymbio, Commerce (BigCommerce & Feedonomics), and Shopware") and a merchant-control promise: with Store Sync "you stay in charge of your business. You control your brand appearance and customer communications for all AI-powered transactions".

Built-in AI = 14.4 ÷ 90 × 100 = 16.0

Automation0.0/100×0.15 of the PA blend

Depth of automation primitives — rules, scheduling, bulk operations, webhooks.

Perform bulk operations across many items at onceweight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/agent-toolsPost-purchase operations are first-class agent tools: refunds via create_refund / get_refund (a "refund for a captured payment", where "If not specified, a refund is for the full amount of the purchase"), disputes via list_disputes ("Retrieve a summary of all disputes with optional filtering"), get_dispute, and accept_dispute_claim ("Accept a dispute claim, resolving it in favor of the buyer."), and orders via create_order ("Create an order in the PayPal system based on the provided details."), pay_order, and create_shipment. Catalog-management tools (create_product, list_product, show_product_details) let agents write to the PayPal product catalog itself.
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [claimed-docs] https://developer.paypal.com/store-sync/create-catalogCatalog publishing is feed-file based: "PayPal supports 3 feed specifications", including the Google Product Feed ("you can reuse it to power both traditional and AI-driven discovery with minimal extra work") and the "OpenAI ACP Product Feed", delivered as CSV/TSV/PSV where "each row in your feed represents a single product variant" and "Your feed file must not exceed 4 GB." Required Google-spec fields include id, title, link, image_link, description ("minimum 25 characters"), price, and availability; beyond the access gate, "access approval is the only prerequisite."

Define rules that trigger actions automatically on eventsweight 3

3 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 30 max

  • [claimed-docs] https://developer.paypal.com/api/webhooks/overview"Webhooks are HTTPS posts from PayPal to an endpoint on your server whenever the corresponding event type occurs" — "Up to 10 webhook URLs may be subscribed per app", each scoped to specific event types or all of them. Delivery is durable: "Any non-2xx status code will cause PayPal to reattempt delivery up to 25 times over the course of 3 days or until it receives a 2xx success code." Authenticity checks are documented ("Post the message, stored webhook ID, and header information back to PayPal's verify signature endpoint"), and the warning is explicit: "If you do not verify a message, you will not have a way to validate that its sender was in fact PayPal."

Schedule recurring jobs or workflowsweight 2

2 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 20 max

  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkit"PayPal's agent toolkit supports the integration of PayPal APIs into AI agent workflows using" Amazon Bedrock, CrewAI, LangChain, MCP, OpenAI's Agents SDK, and Vercel's AI SDK, in TypeScript and Python. Coverage: "Integrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions." The docs' own operating guidance keeps a human in the loop: "Keep human oversight in areas where personal judgment matters most, while letting AI handle repetitive tasks."
  • [claimed-docs] https://developer.paypal.com/ai-tools/agent-toolsPost-purchase operations are first-class agent tools: refunds via create_refund / get_refund (a "refund for a captured payment", where "If not specified, a refund is for the full amount of the purchase"), disputes via list_disputes ("Retrieve a summary of all disputes with optional filtering"), get_dispute, and accept_dispute_claim ("Accept a dispute claim, resolving it in favor of the buyer."), and orders via create_order ("Create an order in the PayPal system based on the provided details."), pay_order, and create_shipment. Catalog-management tools (create_product, list_product, show_product_details) let agents write to the PayPal product catalog itself.
  • [claimed-docs] https://developer.paypal.com/ai-tools/toolkitIntegrate with PayPal APIs to access orders, invoices, subscriptions, shipment tracking, transaction details, and dispute management through pre-built functions.

Version, review, and roll back my automationsweight 1

1 (weight) × 0 (quality) × 0.0 (none) = 0.0 of 10 max

no evidence cited — the verdict rests on absence of evidence, re-checked on refresh

Automation = 0.0 ÷ 80 × 100 = 0.0