cursorrules

Instructions for treating a repository as an MCP server that provides searchable 5eTools data for tabletop role-playing games. The rules also describe how to keep source data unchanged and produce stable links to game entities.

In plain words
What is it for?
Use them when adding search, retrieval, rendering, or documentation for 5eTools entities, including Dungeons & Dragons rules and related content.
Why use it?
They give the agent boundaries for working with read-only game data and consistent rules-version and rendering choices.

Cursor rule for Cursor

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/mrg4dget/5e-tools-mcp/cursorrules
Clone the repo
git clone --depth 1 https://github.com/MrG4dget/5e-tools-mcp

Made for: Cursor.

Per session 485 This file is loaded in full into every session.
When invoked 485 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin 88% copy Near-identical to another mod 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 $0.00485 $0.00485
Opus 5 $0.00243 $0.00243
Sonnet 5 $0.00097 $0.00097
Haiku 4.5 $0.00049 $0.00049

Measured 2d ago against content hash 91a0c3c6eeec, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 2d 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.

Origin

This is a copy

88% identical to mcp-5etools-core — 23 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.cursorrules · 42 lines

What it actually says

Cursor Rules — mcp-5etools

Purpose

Cursor, treat this repository as an MCP server exposing 5eTools data to LLMs. The 5etools-src/ folder at project root is read-only input. Do not rewrite or reformat upstream data files.

High-level goals

  1. Implement stable MCP tools/resources for search, fetch, and render of 5eTools entities.
  2. Prefer 2024 rules when an entity exists in both 2014 and 2024, unless explicitly requested otherwise.
  3. Ensure all generated text links back to fiveet://entity/{kind}/{source}/{slug} URIs.
  4. Keep rendering deterministic, resolving common {@...} tags.

Architecture boundaries

  • src/server.ts is the entry point; all file I/O and indexing stays under src/.
  • The data path defaults to ./5etools-src. Allow override with FIVETOOLS_SRC_DIR env var.
  • Avoid global state leakage; keep indices in a simple in-memory cache for now.

Tasks to prefer

  • Add more {@tag} handlers in the renderer with unit tests.
  • Add fuzzy search and ranking (e.g., minisearch) with facets.
  • Add list_sources, get_rules_section, resolve_tag tools.
  • Write schema/resource docs (ARCHITECTURE.md) when types grow.

Tasks to avoid

  • Editing anything inside 5etools-src/.
  • Introducing non-deterministic rendering (e.g., random ordering, timestamps).
  • Over-fetching: return resource_link when possible.

Coding standards

  • TypeScript strict mode; no any in public types.
  • Small, composable functions; prefer pure utilities for rendering.
  • Keep dependencies minimal (MCP SDK, zod).

PR/commit hygiene

  • One logical concern per PR.
  • Update README/ARCHITECTURE when changing surfaces.
  • Include quick tests for any new tag handler.

Prompts for in-IDE use

  • When unsure, call a tool. Don’t hallucinate rules text.
  • Prefer 2024 rules unless told otherwise.
  • Always include source abbreviations in results.
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. 2d ago First seen · 42 lines · 485 tokens per session scan A 91a0c3c6eeec

Subscribe to this mod's changes

cursorrules is a cursor rule published in the GitHub repository MrG4dget/5e-tools-mcp (2 stars, last pushed 1y ago), licensed MIT. It adds 485 tokens to every session, about $0.0024 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to mcp-5etools-core, differing in 23 lines, and is treated as a copy.