Delphi Clean Code

A set of concise coding rules for Delphi, including naming, small methods, single-purpose classes, avoiding repetition, and choosing the simplest requested solution.

In plain words
What is it for?
It is for writing, reviewing, or refactoring Delphi classes, methods, variables, parameters, fields, and constants.
Why use it?
It gives an agent clear standards for producing code that is easier to read, review, and maintain. It also discourages unnecessary abstractions and duplicated logic.

Skill for Claude CodeCodex

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 skills/delphicleancode/delphi-spec-kit/clean-code
Any agent
npx skills add delphicleancode/delphi-spec-kit --skill clean-code
Clone the repo
git clone --depth 1 https://github.com/delphicleancode/delphi-spec-kit

Made for: Claude Code, Codex.

Per session 22 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,102 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00022 $0.01102
Opus 5 $0.00011 $0.00551
Sonnet 5 $0.00004 $0.00220
Haiku 4.5 $0.00002 $0.00110

Measured 2d ago against content hash b1fc24c7b951, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

Delphi Clean Code scanned grade B with 1 finding 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.

Strips warnings and disclaimersmediumAnti-refusal

Omitting safety caveats hides risk from the user and is a common jailbreak preamble.

| ✅ **No errors?** | Compiles without warnings? |
Origin

Copies of this mod

1 near-identical copy found in the catalogue:

.claude/skills/clean-code/SKILL.md · 150 lines

How it starts

The opening of the file, as written. The whole thing — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Delphi Clean Code — Skill

CRITICAL SKILL — Be concise, direct and solution-focused.

Fundamental Principles

Principle Rule
SRP A function/class does ONE thing
DRY Don't repeat code — extract and reuse
KISS Simplest solution that works
YAGNI Don't build what wasn't asked for
Boy Scout Leave the code better than you found it

Naming Rules (Pascal Guide)

Element Convention
Variables Reveal intent: LCustomerCount not N
Methods Verb + noun: GetCustomerById not Customer
Booleans Question form: IsActive, HasPermission, CanEdit
Constants SCREAMING_SNAKE: MAX_RETRY_COUNT
Fields Prefix F: FCustomerName
Parameters Prefix A: ACustomerName
Var. locations Prefix L: LCustomer

Rule: If you need a comment to explain a name, rename it.

Method Rules

Rule Description
Short Maximum 20 lines, ideal 5-10
One Thing Do one thing and do it well
One Level One level of abstraction per method
Few Args Maximum 3 arguments, prefer 0-2
No Side Effects Don't mute inputs unexpectedly

Code Structure

Standard Application
Guard Clauses Early returns for edge cases
Flat > Nested Avoid deep nesting (max 2 levels)
Composition Small compound methods
Colocation Related code together

Guard Clauses in Delphi

//❌ BAD — excessive nesting
procedure ProcessOrder(AOrder: TOrder);
begin
  if Assigned(AOrder) then
  begin
    if AOrder.Items.Count > 0 then
    begin
      if AOrder.IsValid then
      begin
        //real logic here
      end;
    end;
  end;
end;

//✅ BOM — guard clauses
procedure ProcessOrder(AOrder: TOrder);
begin
  if not Assigned(AOrder) then
    raise EArgumentNilException.Create('AOrder cannot be nil');
  if AOrder.Items.Count = 0 then
    raise EBusinessRuleException.Create('Order must have items');
  if not AOrder.IsValid then
    raise EValidationException.Create('Order validation failed');

  //real logic here — no nesting
end;

Read the full file on GitHub · 150 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 · 150 lines · 22 tokens per session scan B b1fc24c7b951

Subscribe to this mod's changes

Delphi Clean Code is a skill published in the GitHub repository delphicleancode/delphi-spec-kit (50 stars, last pushed 5mo ago), licensed MIT. It adds 22 tokens to every session and 1,102 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

dmvcframework

Use when writing any server-side feature with DelphiMVCFramework (DMVC) — controllers, routes, REST endpoints, entities, middleware, validation, dependency injection, JWT, SSE, or the server bootstrap. The core skill; the Minimal API, web-app, UI, security and testing skills build on it. Triggers on "DMVCFramework"…

danieleteti/delphi-ai-skills · 131 tokens

dmvcframework-jsonrpc

Use when building or consuming a JSON-RPC 2.0 endpoint with DelphiMVCFramework — publishing a plain Delphi class as remotely callable methods, or calling one from a Delphi client. Triggers on "JSON-RPC", "JSONRPC", "MVCFramework.JSONRPC", "RPC method", "PublishObject", "TMVCJSONRPCController", "TMVCJSONRPCPublisher"…

danieleteti/delphi-ai-skills · 139 tokens

dmvcframework-testing

Use when writing or fixing tests for a DelphiMVCFramework API — integration tests that drive real HTTP against an in-process server, or unit tests around controllers. Covers the DUnitX + in-process IMVCServer + IMVCRESTClient stack, CRUD/auth/authorization test patterns, database fixtures and the console runner.…

danieleteti/delphi-ai-skills · 115 tokens

delphi

Use when writing, reviewing or fixing Delphi / Object Pascal code — any .pas or .dpr unit, the RTL or the VCL — independently of any framework. Covers language level and version gating, inline var, generics, anonymous methods, memory and lifetime (Free, try/finally, interfaces, managed records), strings and encodings…

danieleteti/delphi-ai-skills · 155 tokens

dmvcframework-minimal-api

Use when building a DelphiMVCFramework service with lambda/anonymous-method routes instead of controller classes — Minimal API. Triggers on "minimal API", "MapGet", "MapPost", "MapMethods", "route group", "endpoint filter", "HTTP filter", "TMVCRouteGroup", "AsWeb", "lambda routes", "no controller", "Prefix"…

danieleteti/delphi-ai-skills · 125 tokens

dmvcframework-ui

Use when styling or laying out the HTML of a DMVCFramework web application — adding a page, a nav entry, a card, a form, a toast, a theme toggle, or touching style.css. Covers the presentation layer the DMVCFramework wizard actually generates: Bootstrap 5.3 (loaded from CDN in baselayout), the brand tokens in…

danieleteti/delphi-ai-skills · 143 tokens