mcp-tools

A set of coding rules for building MCP tools in Textrawl. MCP is a standard way for an AI assistant to call software tools, and Textrawl is the project these rules describe.

In plain words
What is it for?
Use it when implementing or reviewing Textrawl MCP tools, including their registration, input validation, structured results, logging, and read-only or destructive behavior.
Why use it?
It keeps tool definitions, inputs, outputs, errors, and safety hints consistent across the project.

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/jeffgreendesign/textrawl/mcp-tools
Clone the repo
git clone --depth 1 https://github.com/jeffgreendesign/textrawl

Made for: Cursor.

Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 1,081 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 $0.00000 $0.01081
Opus 5 $0.00000 $0.00541
Sonnet 5 $0.00000 $0.00216
Haiku 4.5 $0.00000 $0.00108

Measured 2d ago against content hash 8d54d8ae7625, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcp-tools 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/mcp-tools.mdc · 124 lines

How it starts

The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MCP Tool Patterns

Tool Registration (Preferred: registerTool)

Use server.registerTool() (SDK v1.26.0+) with title, description, inputSchema, outputSchema, and annotations:

import type { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js';
import { z } from 'zod';
import { configError, formatId, isCompact, toolError, toolResponse } from '../utils/compact.js';
import { logger } from '../utils/logger.js';

server.registerTool(
  'tool_name',
  {
    title: 'Tool Name',
    description: 'What this tool does',
    inputSchema: {
      query: z.string().min(1).max(10000).describe('Search query'),
      limit: z.number().min(1).max(50).default(10).describe('Max results'),
    },
    // For read-only tools: define outputSchema + return structuredContent
    outputSchema: {
      totalResults: z.number(),
      results: z.array(z.object({ id: z.string(), content: z.string() })),
    },
    annotations: {
      readOnlyHint: true,     // Does not modify data
      destructiveHint: false,  // Does not delete data
      idempotentHint: true,    // Safe to call multiple times
      openWorldHint: false,    // Does not make external API calls
    },
  },
  async ({ query, limit }) => {
    logger.info('tool_name called', { query, limit });

    if (!isSupabaseConfigured()) {
      return configError('Database', 'Set SUPABASE_URL and SUPABASE_SERVICE_KEY');
    }

    try {
      const results = await doQuery(query, limit);
      const structuredContent = { totalResults: results.length, results };

      return toolResponse({
        compact: { n: results.length, r: results.map(r => ({ id: formatId(r.id), c: r.content })) },
        verbose: structuredContent,
        structuredContent,  // Always verbose, canonical keys
      });
    } catch (error) {
      // Structured form — preferred. Logging happens inside toolError().
      return toolError('tool_name', error, { scope: 'query' });
    }
  }
);

logger.debug('Registered tool: tool_name');

Read the full file on GitHub · 124 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 · 124 lines · 0 tokens per session scan A 8d54d8ae7625

Subscribe to this mod's changes

mcp-tools is a cursor rule published in the GitHub repository jeffgreendesign/textrawl (5 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,081 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 2026-08-31.