architecture

An architecture guide for Ditto MCP, a server that runs Ditto DQL queries through an MCP connection. It defines where the command-line entrypoint, server tools, configuration, HTTP client, safety checks, and logging belong.

In plain words
What is it for?
It guides changes to server startup, tool registration, environment settings, Ditto HTTP requests, allowed query operations, timeouts, error handling, graceful shutdown, and leveled logging.
Why use it?
It keeps modules within clear boundaries and ensures database queries are restricted and sensitive information is redacted.

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/evtapps/ditto-mcp/architecture
Clone the repo
git clone --depth 1 https://github.com/evtapps/ditto-mcp

Made for: Cursor.

Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 530 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 $0.00016 $0.00530
Opus 5 $0.00008 $0.00265
Sonnet 5 $0.00003 $0.00106
Haiku 4.5 $0.00002 $0.00053

Measured yesterday against content hash 7e09a25a3af1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

architecture 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.

.cursor/rules/architecture.mdc · 37 lines

What it actually says

Ditto MCP – Architecture Guide

This repository implements an MCP (Model Context Protocol) server that executes Ditto DQL with safety guardrails.

  • Entrypoint CLI: src/index.ts creates the server and connects a stdio transport.
  • Server factory: src/mcpServer.ts registers tools/resources. It is the single place that adds tools to the MCP list.
  • Config: src/config.ts parses environment variables and CLI overrides and returns a typed ServerConfig.
  • Ditto HTTP client: src/ditto/client.ts contains the executeDql function.
  • Guardrails: src/guards/dql.ts enforces single-statement, operation allow-list, and optional regex allow-list.
  • Logging: src/utils/logger.ts handles leveled logging and redaction.

Responsibilities and Boundaries

  • src/index.ts: CLI parsing, process error handling, transport lifecycle, graceful shutdown.
  • src/mcpServer.ts: Tool and resource definitions only. No direct env access; uses values passed via ServerConfig.
  • src/config.ts: Source of truth for allowed operations and timeouts. Only this module reads env variables.
  • src/ditto/*: Pure HTTP logic. It returns typed results or a Ditto-style error envelope.
  • src/guards/*: Stateless validation/guard code. Server must call before any network I/O.
  • dist/*: Generated output. Never edit.

Transport

  • Current transport is stdio. The CLI argument chooses transport string, but server connects stdio by default.
  • HTTP transport can be added later as an orthogonal module; keep mcpServer.ts free of transport details.

Data Flow

  • CLI + env → loadConfig()ServerConfigcreateServer() → tool handler → guard functions → Ditto HTTP client → typed response → MCP content.

Error Handling

  • Process-level handlers in src/index.ts convert unknown errors to strings safely and exit on uncaughtException.
  • Tool handlers return { isError: true, content: [...] } when validation or remote errors occur.
  • HTTP client returns a Ditto-like envelope on failure to simplify tool logic.
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 · 37 lines · 16 tokens per session scan A 7e09a25a3af1

Subscribe to this mod's changes

architecture is a cursor rule published in the GitHub repository evtapps/ditto-mcp (0 stars, last pushed 9mo ago), licensed MIT. It adds 16 tokens to every session and 530 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.