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.
git clone --depth 1 https://github.com/beel-collab/presets.devWrote 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/beel-collab/presets.dev/mcp-rules)<a href="https://agentmods.dev/rules/beel-collab/presets.dev/mcp-rules"><img src="https://agentmods.dev/badge/rules/beel-collab/presets.dev/mcp-rules/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/beel-collab/presets.dev/mcp-rules"><img src="https://agentmods.dev/badge/rules/beel-collab/presets.dev/mcp-rules.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00000 | $0.01872 |
| Opus 5.5 | $0.00000 | $0.00749 |
| Sonnet 5.5 | $0.00000 | $0.00374 |
| Haiku 4.5 | $0.00000 | $0.00187 |
Grade A, and why
mcp-rules 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 today.
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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MCP Rules for Cursor: Tools, Schemas and Permissions
Install
Install the native export at .cursor/rules/mcp-rules.mdc.
Copy that export, not this catalog file. Catalog fields such as tools and author are for presets.dev.
Install one skill file per name. Cursor, GitHub Copilot, Codex, Antigravity CLI, and Devin can all read .agents/skills, so a second copy with the same name collides.
MCP Rules for Cursor: Tools, Schemas and Permissions
File-scoped Cursor rule for MCP. It attaches when a matching file is in context.
Updated for the 2026-07-28 MCP spec, Nov 2026
Do not upgrade dependencies or the framework version unless the task asks.
The spec this file follows is the 2026-07-28 revision. @modelcontextprotocol/sdk 1.32.2 is still the v1 package on npm, and a server that calls server.connect(new StdioServerTransport()) or hands requests to a Streamable HTTP transport from that package speaks the 2025-era protocol only. A client pinned to 2026-07-28 cannot complete a session with it.
To serve 2026-07-28, use the v2 packages. @modelcontextprotocol/server 2.4.0 and @modelcontextprotocol/client 2.4.0 are the v2 packages that speak this revision. Import McpServer and createMcpHandler from @modelcontextprotocol/server. Import serveStdio from @modelcontextprotocol/server/stdio. Pass a factory that constructs a new McpServer on each call. A handler that receives Node IncomingMessage imports its HTTP transport from @modelcontextprotocol/node. Read the installed package names before editing. Leave a v1 server on v1 unless the task is the move to v2, and do not import @modelcontextprotocol/core-internal.
What this revision removed
The 2026-07-28 revision is stateless. There is no initialize / notifications/initialized handshake and no Mcp-Session-Id header. Every request carries its protocol version and client capabilities in _meta (io.modelcontextprotocol/protocolVersion, io.modelcontextprotocol/clientCapabilities). Servers identify themselves on each result the same way. A version mismatch is UnsupportedProtocolVersionError.
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.
- today First seen · 132 lines · 0 tokens per session scan A 7a106da71e0c
mcp-rules is a cursor rule published in the GitHub repository beel-collab/presets.dev (3 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,872 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 2026-11-13.
mcp-rules compared with cursor rules from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| mcp-rulesbeel-collab/presets.dev | Cursor rule | 3 | yesterday | MIT | A | 0 | Cursor |
| ponytailDietrichGebert/ponytail | Cursor rule | 190k | 15d ago | MIT | A | 604 | Cursor |
| angular-20angular/angular | Cursor rule | 101k | today | MIT | A | 0 | Cursor |
| dev-standardapache/superset | Cursor rule | 76k | today | Apache-2.0 | A | 1,147 | Cursor |
| prefer-direct-imports-over-module-mocksprisma/orm | Cursor rule | 48k | today | Apache-2.0 | A | 0 | Cursor |
| config-validation-and-normalizationprisma/orm | Cursor rule | 48k | today | Apache-2.0 | A | 0 | Cursor |
| control-plane-descriptorsprisma/orm | Cursor rule | 48k | today | Apache-2.0 | A | 0 | Cursor |