new-tool

A step-by-step set of rules for adding a new MCP tool to a project. MCP tools are actions an AI agent can call, such as archiving an email.

In plain words
What is it for?
Creating a tool handler, registering it with its name and input schema, calling Microsoft Graph when needed, and formatting the result.
Why use it?
It turns adding a tool into a repeatable process and keeps handlers, registration, error handling, and responses consistent.

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/titanzero/outlook-mcp/new-tool
Clone the repo
git clone --depth 1 https://github.com/titanzero/outlook-mcp

Made for: Cursor.

Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 553 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin fork From a forked repository.
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 $0.00000 $0.00553
Opus 5 $0.00000 $0.00277
Sonnet 5 $0.00000 $0.00111
Haiku 4.5 $0.00000 $0.00055

Measured 2d ago against content hash c9c4fa1499d7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

new-tool 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.

.cursor/rules/new-tool.mdc · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Adding a New MCP Tool

Steps

1. Create the handler file

Create <module>/<action>.js (example: email/archive.js) and keep the handler focused on business logic plus Graph calls:

const { getGraphClient } = require('../utils/graph-client');
const { formatResponse } = require('../utils/response-formatter');
const { makeResponse } = require('../utils/response-helpers');

async function handleArchiveEmail(args) {
  const client = await getGraphClient();
  // ... Graph API call ...
  const text = formatResponse(structuredData, textFallback);
  return makeResponse(text);
}

module.exports = handleArchiveEmail;

Follow core conventions from general.mdc for error handling and shared response patterns.

2. Register in module index

Edit <module>/index.js:

const handleArchiveEmail = require('./archive');

const emailTools = [
  // ... existing tools ...
  {
    name: "archive-email",
    description: "Moves an email to the archive folder",
    inputSchema: {
      type: "object",
      properties: {
        id: { type: "string", description: "Email ID to archive" }
      },
      required: ["id"]
    },
    handler: handleArchiveEmail
  }
];

Use the module's {module}Tools array (emailTools, calendarTools, etc.). The root index.js aggregates these arrays into TOOLS.

3. Add config fields if needed

If the tool needs new Graph API field selections or constants, add them to config.js.

4. Write tests

Create test/<module>/<action>.test.js following the patterns in the testing rule.

5. Update CLAUDE.md

Update docs when the tool adds new capabilities or config requirements.

Checklist

  • Handler file exports one async handler function
  • Handler uses getGraphClient() and Graph SDK fluent calls
  • Tool registered in <module>/index.js under {module}Tools with name, description, inputSchema, handler
  • New constants/field sets are added in config.js when required
  • Test file created in test/<module>/ (follow testing.mdc)
  • Core conventions in general.mdc are followed

Read the full file on GitHub · 75 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. 2d ago First seen · 75 lines · 0 tokens per session scan A c9c4fa1499d7

Subscribe to this mod's changes

new-tool is a cursor rule published in the GitHub repository titanzero/outlook-mcp (1 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 553 tokens. A static security scan graded it A with 0 findings. It comes from a forked repository.