Head to head · Checkout · October 2026 research run

Adyen MCP server vs Paid

Adyen MCP server scores 60.3 (C) on agent readiness against Paid's 55.1 (C), and leads in 4 of 7 scored categories. Paid leads on agent ergonomics and maintenance & community. Both do checkout.

Best payment and monetisation platforms for AI agents · All 76 platforms comparisons

Which one, for what

Adyen MCP server C

Good for An Adyen merchant's developer or operations staff who want an assistant to create payment links and sessions, look up terminals, webhooks, users and credentials, and change terminal settings, with a person watching writes.

Ahead on

  • Reliability, 65 against 35
  • Schema & documentation, 73 against 64
  • Payments & pricing, 35 against 30
  • Transparency & trust, 83 against 51

Also in its favour

  • Runs on your own machine
  • Open source

Watch for

The README is headed alpha, the package is at 0.6.0, and the server reports itself as version 0.4.0

Paid C

Good for A team selling an AI agent that needs usage metering, credit balances, cost and margin tracking and invoicing in one service, with Stripe as the payment rail.

Ahead on

  • Agent ergonomics, 70 against 56
  • Maintenance & community, 77 against 42

Also in its favour

  • A hosted endpoint, with nothing to install
  • Free to start without a card

Watch for

The docs index advertises an OpenAPI 3.1 file at two addresses, and both returned 404 on 9 October 2026

Score by category

CategoryWeight this runAdyen MCP serverPaidEdge
Reliability16%206535Adyen MCP server +30
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.27364Adyen MCP server +9
Agent ergonomics13%16.25670Paid +14
Security & auth14%17.56365Paid +2
Payments & pricing10%12.53530Adyen MCP server +5
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.84277Paid +35
Transparency & trust7%8.88351Adyen MCP server +32
Negative events≤1500
Total60.3 · C55.1 · C

Facts side by side

FactAdyen MCP serverPaid
KindMCP serverHTTP API
VendorAdyen N.V.Agent Paid Limited
Hosted endpointno (local only)https://api.agentpaid.io/api/v2
TransportsstdioHTTP, Streamable HTTP
AuthAPI keyOAuth or key
PricingPay per useFreemium
x402nono
LicenceMIT for the server. The Adyen APIs it calls are a closed service under a merchant agreement and the Adyen Terms and ConditionsProprietary service under Paid's Master Subscription Agreement. The Node and Python SDKs on GitHub are MIT
Tools exposed38none
Read-only variant documentedyesyes
llms.txtyesyes
Last release2026-08-242026-09-15
Terms last updated2023-11-01no date given
Privacy policy last updated2025-08-04no date given
Customer content may train modelsyesnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waiveryesnot found in the text
Popularity27 stars, 168 npm/wk24k npm/wk

Verdicts

Adyen MCP server

All 38 tools carry read-only, destructive and idempotent annotations, and --tools loads a subset under an API credential limited by roles. The README labels the server alpha, refund_payment and the other POST tools send no idempotency key, and failures come back as plain text without an error flag.

Paid

Paid suits a builder who wants usage metering, credits and invoicing behind one API, with restricted and customer-bound keys, OAuth on the MCP server and a free plan that needs no card. No status page, SLA figure, API changelog or rate-limit numbers were found, both advertised OpenAPI links returned 404, and card payments need the builder's own Stripe account.

Before you call either

Adyen MCP server

  1. Set ADYEN_API_KEY in the environment and start with npx -y @adyen/mcp --env=TEST. The --adyenApiKey flag in the docs page puts the key in the process list
  2. Pass --tools= with only the tool names the task needs. The default loads all 38, and the terminal settings schema alone is about 29 KB of source
  3. Call refund_payment with pspReference, currency, value, merchantAccount and reference. The description's paymentPspReference is not the field name
  4. Treat a Failed to ... reply from a write tool as outcome unknown, not as failure. Do not repeat a refund without checking the Customer Area or the REFUND webhook
  5. Give amounts in minor units (1099 for 10.99 EUR). Live use needs --env=LIVE --livePrefix= with the prefix from the live Customer Area

Paid

  1. Call https://api.agentpaid.io/api/v2 with Authorization: Bearer <key>. A person creates the first key at app.paid.ai. No programmatic signup was found
  2. Give every signal a stable idempotencyKey and resend the same key on retry. Signals without one are not deduplicated
  3. Send at most 500 signals a request. A 429 rejects the whole batch, so wait for Retry-After and resend it all
  4. Read the code field on a 429. CONCURRENCY_LIMIT_EXCEEDED carries no Retry-After and clears when an analytics query finishes
  5. Install the Python SDK as paid-python, the name in its README and pyproject.toml. The quickstart's pip install paid names a package we could not check

Questions

Which is better for AI agents, Adyen MCP server or Paid?

Adyen MCP server scores 60.3 (C) on agent readiness against Paid's 55.1 (C), and leads in 4 of 7 scored categories. Paid leads on agent ergonomics and maintenance & community.

Do Adyen MCP server and Paid need an API key?

Adyen MCP server needs an API key. Paid takes an API key or an OAuth sign-in.

Can an agent call Adyen MCP server and Paid without installing anything?

Adyen MCP server runs on your own machine, with no hosted endpoint listed. Paid has a hosted endpoint at https://api.agentpaid.io/api/v2.

Are Adyen MCP server and Paid open source?

Adyen MCP server is open source (MIT for the server. The Adyen APIs it calls are a closed service under a merchant agreement and the Adyen Terms and Conditions). No open-source release is listed for Paid.

Other comparisons with Adyen MCP server or Paid

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.