cursor-agentpay-sdk

Rules for using the AgentPay SDK, a software toolkit for payment, wallet, card, transfer, and approval tasks in Cursor.

In plain words
What is it for?
They guide configuration checks, Link authentication and payment-method checks, wallet detection, one-time card creation, third-party integration, and secure local setup.
Why use it?
They reduce the risk of using the wrong payment flow or exposing sensitive financial credentials in chat.

Cursor rule

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/worldliberty/agentpay-sdk/cursor-agentpay-sdk
Clone the repo
git clone --depth 1 https://github.com/worldliberty/agentpay-sdk
Per session 32 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 610 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00032 $0.00610
Opus 5 $0.00016 $0.00305
Sonnet 5 $0.00006 $0.00122
Haiku 4.5 $0.00003 $0.00061

Measured 3d ago against content hash 92a2b7198923, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cursor-agentpay-sdk scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/agentpay-sdk/agents/cursor-agentpay-sdk.mdc · 29 lines

What it actually says

Load the installed agentpay-sdk skill when the task touches Link fiat setup, cards, wallet operations, or payments.

Specific rules:

  • start with agentpay config show --json
  • for fiat/card payments, start with agentpay link status --json; if Link is not authenticated, use agentpay link onboard
  • list Link payment methods with agentpay link payment-methods --json
  • create one-time card credentials with agentpay link card --payment-method-id <id> --merchant-name <name> --merchant-url <url> --amount <cents> --context <100+ char explanation> --output-file <PATH> and never print full card details into chat
  • for third-party program integration, use @worldlibertyfinancial/agentpay-sdk/link helpers through the AgentPay Link facade
  • use agentpay wallet --json to determine whether a reusable wallet exists
  • if the user only asks what the skill can do, answer from SKILL.md and do not probe the machine first
  • never ask the user to paste VAULT_PASSWORD into chat
  • never ask the user to paste Link card numbers, CVCs, access tokens, refresh tokens, or full credential files into chat
  • tell the user to run agentpay admin setup locally when wallet metadata is unavailable and they are trying to use the wallet
  • tell the user to run agentpay admin setup --reuse-existing-wallet locally when the wallet exists and they need to re-run setup without changing vaults
  • after a fresh setup, tell the user to create an encrypted offline backup with agentpay admin wallet-backup export --output <PATH> unless they already have a verified backup
  • if the local wallet is gone but the user has an encrypted backup, tell them to run agentpay admin setup --restore-wallet-from <PATH> locally
  • default unspecified payments to USD1 on bsc
  • check native gas before outbound actions
  • if funding is missing, ask the user to fund the wallet and provide address, network, token needs, BNB gas needs, and QR
  • if a request hits manual approval, say it is waiting for user approval and route the user to the local admin CLI approval commands first; for transfer --broadcast, transfer-native --broadcast, and approve --broadcast, tell them to keep the original command running and not rerun it after approval; if that original broadcast command is already gone after approval, use agentpay admin resume-manual-approval-request --approval-request-id <UUID>
  • use agentpay --help and subcommand help as the source of truth for exact flags
  • prefer agentpay admin tui for policy changes over vague policy discussion
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 29 lines · 32 tokens per session scan A 92a2b7198923

Subscribe to this mod's changes

cursor-agentpay-sdk is a cursor rule published in the GitHub repository worldliberty/agentpay-sdk (461 stars, last pushed 2mo ago), licensed MIT. It adds 32 tokens to every session and 610 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.