add-tool

A command that creates the initial TypeScript file for a new MCP tool, including input and output schemas and a placeholder handler. MCP is a standard way for an AI application to call external tools.

In plain words
What is it for?
Use it when adding a tool to an example connector or to your own project. Provide a valid tool name, then fill in its schemas and handler.
Why use it?
It gives a new tool the expected project structure and validation setup, so you do not have to create the boilerplate by hand.

Command for Claude Code

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 commands/lelemon-studio/mcify/add-tool
Clone the repo
git clone --depth 1 https://github.com/Lelemon-studio/mcify

Made for: Claude Code.

Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 584 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00015 $0.00584
Opus 5 $0.00008 $0.00292
Sonnet 5 $0.00003 $0.00117
Haiku 4.5 $0.00002 $0.00058

Measured 2d ago against content hash 6829de0a0954, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

add-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.

.claude/commands/add-tool.md · 68 lines

What it actually says

Scaffold a new MCP tool — either in an example connector or in the user's own project.

Required input from the user: the tool name (e.g., khipu_create_payment).

  1. Validate the name against /^[a-zA-Z0-9_-]{1,64}$/. Reject if it doesn't match (this is the same regex defineTool enforces at runtime).

  2. Find the right home for the tool. Ask the user only if it's not obvious:

    • In a packages/examples/<connector>/ example: add to that connector's tools/ directory.
    • In a user project: add to tools/ next to mcify.config.ts.
  3. Write the tool file as <name>.ts:

    import { defineTool, schema } from '@mcify/core';
    import { z } from 'zod';
    
    export const <camelName> = defineTool({
      name: '<name>',
      description: '<one sentence, present tense>',
      input: z.object({
        // schema.id(), schema.url(), schema.money(), etc. for common shapes
      }),
      output: z.object({
        // ...
      }),
      handler: async (input, ctx) => {
        throw new Error('not implemented');
      },
    });
    

    Suggest sensible Zod helpers from @mcify/core's schema namespace where applicable (schema.id, schema.url, schema.httpUrl, schema.money, schema.timestamp, schema.paginated).

  4. Register it in mcify.config.ts:

    import { <camelName> } from './tools/<name>.js';
    
    export default defineConfig({
      // ...
      tools: [<camelName>, /* ... existing ... */],
    });
    
  5. Add a basic test as tools/<name>.test.ts:

    import { describe, it, expect } from 'vitest';
    import { <camelName> } from './<name>.js';
    
    describe('<name>', () => {
      it('rejects invalid input', () => {
        const result = <camelName>.input.safeParse({});
        expect(result.success).toBe(false);
      });
    
      // Add a happy-path test once the handler is implemented.
    });
    
  6. Verify: run pnpm typecheck and pnpm test for the affected package. Don't write the actual handler logic — the developer fills that in.

  7. Don't add a changeset unless the tool ships in a published example. User-project tools never get changesets.

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 · 68 lines · 15 tokens per session scan A 6829de0a0954

Subscribe to this mod's changes

add-tool is a command published in the GitHub repository Lelemon-studio/mcify (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 15 tokens to every session and 584 once invoked, about $0.0001 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 2026-08-31.