cursorrules

A set of Cursor rules for organising tools in a Directus Model Context Protocol server. Directus is a backend platform for managing data and content, while toolsets group related operations.

In plain words
What is it for?
Use it when adding schema, content, or workflow tools and deciding which toolsets should expose them.
Why use it?
It prevents new tools from being placed in the wrong group and protects the expected default behaviour.

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/skeyelab/directus-mcp-server/cursorrules
Clone the repo
git clone --depth 1 https://github.com/Skeyelab/directus-mcp-server

Made for: Cursor.

Per session 506 This file is loaded in full into every session.
When invoked 506 The same file — it is already loaded in full.
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 $0.00506 $0.00506
Opus 5 $0.00253 $0.00253
Sonnet 5 $0.00101 $0.00101
Haiku 4.5 $0.00051 $0.00051

Measured yesterday against content hash 67166ab16c97, 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 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.

.cursorrules · 49 lines

How it starts

The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Directus MCP Server Cursor Rules

Toolset Organization

When adding new tools or features to this MCP server, always organize them into appropriate toolsets:

  1. Schema and Content tools must belong to the default toolset - This ensures backward compatibility and allows clients to access core tools when no specific toolset is requested.

  2. Flow tools do NOT belong to the default toolset - Flow tools are only in the flow toolset and must be explicitly requested.

  3. Assign tools to their specific toolset:

    • Schema-related tools → schema toolset (collections, fields, relations, schema snapshots)
    • Content-related tools → content toolset (items CRUD, queries, bulk operations)
    • Flow-related tools → flow toolset (workflow automation, triggers) - NOT in default
  4. Toolset assignment format:

    // For schema and content tools:
    {
      name: 'tool_name',
      description: 'Tool description',
      inputSchema: z.object({...}),
      toolsets: ['default', 'schema'] as const,  // Always include 'default' + specific toolset
      handler: async (client, args) => {...}
    }
    
    // For flow tools:
    {
      name: 'tool_name',
      description: 'Tool description',
      inputSchema: z.object({...}),
      toolsets: ['flow'] as const,  // Only 'flow', NOT 'default'
      handler: async (client, args) => {...}
    }
    
  5. When creating new tool files:

    • Determine which toolset the tools belong to based on their functionality
    • Schema and content tools: Add to both default and their specific toolset
    • Flow tools: Add ONLY to flow toolset (not default)
    • Update the toolset filtering logic in src/index.ts if adding a new toolset category
  6. Toolset types are defined in src/types/index.ts as:

    export type Toolset = 'default' | 'schema' | 'content' | 'flow';
    

This organization allows clients to filter tools using the MCP_TOOLSETS environment variable, improving AI agent accuracy by reducing tool confusion. The default toolset contains only schema and content tools, while flow tools must be explicitly requested.

Read the full file on GitHub · 49 lines

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. yesterday First seen · 49 lines · 506 tokens per session scan A 67166ab16c97

Subscribe to this mod's changes

cursorrules is a cursor rule published in the GitHub repository Skeyelab/directus-mcp-server (0 stars, last pushed 9mo ago), licensed MIT. It adds 506 tokens to every session, about $0.0025 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 2026-09-01.