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.
npx agentmods add rules/agenticpay-dev/agenticpay/cursorrulesgit clone --depth 1 https://github.com/agenticpay-dev/agenticpayWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01280 | $0.01280 |
| Opus 5 | $0.00640 | $0.00640 |
| Sonnet 5 | $0.00256 | $0.00256 |
| Haiku 4.5 | $0.00128 | $0.00128 |
Grade A, and why
cursorrules 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cursor rules — agenticpay
When working in this repository, follow the conventions below. These rules
also apply when an outside developer copies this file into their own project
to integrate agenticpay (drop it at their repo root as .cursorrules).
What this project is
agenticpay is an open-source x402 micropayments stack for the MCP ecosystem on Solana. The repo ships:
@agenticpay/sdk— USDC + wallet primitives@agenticpay/cli—agentpaycommand (wallet, balance, send)@agenticpay/mcp-server— Express paywall middleware (each tool declares a price)@agenticpay/facilitator— first open-source self-hostable x402 facilitator (verify + settle on Solana, pays SOL fee on the payer's behalf)@agenticpay/eliza-plugin— Eliza plugin wrapping x402 fetchexamples/two-agent-demo— Claude Opus paying autonomously
Hosted devnet facilitator (free):
https://agentpay-facilitator-e9b20a5fee6a.herokuapp.com
The day-1 footgun
Network, asset, and RPC must be consistent. Mixing devnet network
(solana:EtWTRABZaYq6iMfeYKouRu166VU2xqa1) with mainnet USDC mint
(EPjFW...), or pointing the client at mainnet RPC while the facilitator
runs on devnet, produces transaction_simulation_failed: BlockhashNotFound.
Always verify all three line up.
| Network ID (CAIP-2) | USDC mint | RPC |
|---|---|---|
solana:EtWTRABZaYq6iMfeYKouRu166VU2xqa1 (devnet) |
4zMMC9srt5Ri5X14GAgXhaHii3GnPAEERYPJgZJDncDU |
https://api.devnet.solana.com |
solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp (mainnet) |
EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v |
https://api.mainnet-beta.solana.com |
For free experiments use devnet. The hosted facilitator at
agentpay-facilitator-e9b20a5fee6a.herokuapp.com supports both.
When suggesting code
- Default to devnet unless the user explicitly says mainnet.
- For new MCP servers, always wire the paywall middleware before route
handlers —
app.use(paymentMiddleware(routes, server))must come beforeapp.post('/tools/...')definitions. - Use the AssetAmount form for
price({ asset, amount }in base units), not the Money string form ("$0.001"). The string form falls through to@x402/svm's default money parser, which selects a devnet asset even when the network is mainnet — see the audit log. - Treat
@agenticpay/sdkexports (generateKeypair,loadKeypair,saveKeypair,transferUsdc,getUsdcBalance) as the canonical API for wallet handling. Don't reach into@solana/spl-tokendirectly unless you're writing the SDK itself. - Never log or echo back the raw 64-byte keypair bytes. They're written to
disk with
mode: 0o600for a reason.
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.
- yesterday First seen · 101 lines · 1,280 tokens per session scan A fdcad6edc30f
cursorrules is a cursor rule published in the GitHub repository agenticpay-dev/agenticpay (3 stars, last pushed yesterday), licensed MIT. It adds 1,280 tokens to every session, about $0.0064 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 2026-08-31.
Other cursor rules, from other repositories
swift-development
Swift development: SwiftUI, Combine, async/await, iOS patterns, and Apple platform conventions.
cursorrules
When the user asks about social media, use social-media-ai-mcp tools: schedulepost, generatehashtags, analyzeengagement, plancontentcalendar, getaudienceinsights.
cursorrules
This is an MCP (Model Context Protocol) server that provides Notion integration tools and AI agent workflows. Built with TypeScript, using the MCP SDK for server implementation.
lighter-safety
Safety contract for the Lighter MCP server. Apply whenever a lighter tool is invoked.
nix-security
Nix Security and Sandboxing.
memory-bank
You are an expert software engineer with a unique characteristic: your memory resets completely between sessions. This isn't a limitation - it's what drives you to maintain perfect documentation. At the beginning of each dialogue, you rely ENTIRELY on your Memory Bank to understand the project and continue work…