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/solvapay/solvapay-sdk/mcp-apps-sdkgit clone --depth 1 https://github.com/solvapay/solvapay-sdkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/solvapay/solvapay-sdk/mcp-apps-sdk)<a href="https://agentmods.dev/rules/solvapay/solvapay-sdk/mcp-apps-sdk"><img src="https://agentmods.dev/badge/rules/solvapay/solvapay-sdk/mcp-apps-sdk.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.02196 |
| Opus 5 | $0.00000 | $0.01098 |
| Sonnet 5 | $0.00000 | $0.00439 |
| Haiku 4.5 | $0.00000 | $0.00220 |
Grade A, and why
mcp-apps-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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SolvaPay MCP Apps SDK — rules
Rules for building and refactoring the SolvaPay SDK for MCP Apps. Read before writing any code in packages/mcp-core, packages/mcp, packages/mcp-express, packages/mcp-fetch, packages/react/mcp, or examples/mcp-checkout-app / examples/supabase-edge-mcp.
Full contributor-facing copy: docs/contributing/mcp-apps-sdk-rules.md.
North Star
The merchant's data is the hero. Commerce defers to it.
Every design decision should serve this. When in doubt, ask: does this make the merchant's tool feel more native to Claude/ChatGPT, or less? If less, don't ship it.
The three modes
Every SolvaPay response is in one of three modes. Know which before writing code.
- Silent. Merchant's tool returned data. No iframe, no card, no upsell. Just the data. This is 90% of calls for a paying user.
- Nudge. Data returned and something is worth flagging (low balance, cycle ending, approaching limit). Small inline strip. Dismissible. Never blocks.
- Gate. Data could not be returned. User is out of credits or needs to upgrade. SDK takes over the surface. Focused, terminal, collapses after.
If you're building a UI surface that doesn't match one of these three, stop and escalate.
Rules
Commerce UI
- Do not add tabs to the MCP shell. The four-tab shell is deprecated. Each intent (
/manage_account,/topup,/upgrade) opens a single-purpose surface and returns to chat when done. - Do not build an About surface. Product description lives in tool descriptions, Claude's text response, and the
docs://solvapay/overview.mdresource. Not in an iframe. - The UI is a mode, not a primary.
McpAppShellis an internal composition. What's exported as primary is intent-specific surfaces. - One surface, one job. No nested navigation. No multi-step wizards except where genuinely unavoidable (card entry, top-up amount selection). If a surface needs tabs to fit its content, the content is too broad.
- Route on
structuredContent.view, not tool name.<McpApp>resolves the surface from the server-stampedviewdiscriminator (checkout/account/topup) on any intent-tool result. The paywall / nudge surfaces were removed with the text-only paywall refactor — merchant paywall / nudge responses narrate incontent[0].textand don't mount the widget. The iframe opens only when the user (or LLM) deliberately invokes one of the three intent tools (upgrade/manage_account/topup).
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.
- 4d ago First seen · 98 lines · 0 tokens per session scan A 61bb0c6d07e7
mcp-apps-sdk is a cursor rule published in the GitHub repository solvapay/solvapay-sdk (5 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,196 tokens. 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
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.
family-instance-domain-actions
Family instance domain action implementation patterns.