ai-dev-skillset: Skill for Claude Code

.agents/skills/process-tdd/SKILL.md

process-tdd is a skill for Claude Code, Codex from gianverdum/ai-dev-skillset. It costs 175 tokens per session (1,640 once invoked), scanned A, original, MIT.

A test-driven development guide. TDD means writing a failing test for the intended behavior first, then writing the smallest code change that makes it pass, and refactoring afterward.

In plain words
What is it for?
Adding features, fixing bugs, refactoring, and extending software through a repeatable test-first cycle followed by testing, linting, formatting, and compliance checks.
Why use it?
It makes the expected behavior explicit and helps catch regressions while software is being changed.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is gianverdum/ai-dev-skillset's own configuration. It tells Claude Code and Codex how to work on ai-dev-skillset itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything ai-dev-skillset configures →

Reuse

Borrowing it

Nothing to install: this file belongs to gianverdum/ai-dev-skillset. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/gianverdum/ai-dev-skillset/main/.agents/skills/process-tdd/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/gianverdum/ai-dev-skillset

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for process-tdd

README.md
[![agentmods](https://agentmods.dev/badge/skills/gianverdum/ai-dev-skillset/process-tdd/github.svg)](https://agentmods.dev/skills/gianverdum/ai-dev-skillset/process-tdd)
Your own site
<a href="https://agentmods.dev/skills/gianverdum/ai-dev-skillset/process-tdd"><img src="https://agentmods.dev/badge/skills/gianverdum/ai-dev-skillset/process-tdd/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for process-tdd

Your own site · 80×15
<a href="https://agentmods.dev/skills/gianverdum/ai-dev-skillset/process-tdd"><img src="https://agentmods.dev/badge/skills/gianverdum/ai-dev-skillset/process-tdd.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 175 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,640 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00175 $0.01640
Opus 5 $0.00088 $0.00820
Sonnet 5 $0.00035 $0.00328
Haiku 4.5 $0.00017 $0.00164

Measured 9d ago against content hash 9d39397f9552, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

process-tdd 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 9d 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.

.agents/skills/process-tdd/SKILL.md · 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.

TDD

Todo desenvolvimento deve ser guiado por TDD.

Regra Principal

  • Antes de implementar ou alterar comportamento, escreva ou ajuste um teste que falhe pelo motivo esperado.
  • Só escreva código de produção suficiente para fazer o teste passar.
  • Depois que os testes passarem, refatore mantendo a suíte verde.

Ciclo

  1. Entenda o comportamento esperado.
  2. Identifique o menor teste que demonstra esse comportamento.
  3. Escreva o teste primeiro e execute para confirmar a falha.
  4. Implemente a menor mudança suficiente para passar.
  5. Execute os testes relevantes.
  6. Refatore se necessário, mantendo os testes passando.
  7. Repita o ciclo para o próximo comportamento.
  8. Antes de encerrar a tarefa, rode lint e formatter conforme a skill quality-lint-format e corrija o que for apontado.
  9. Como ultima etapa, rode a skill quality-revisao-aderencia para verificar que estrutura, tipagem do dominio, localizacao do parsing, codigos de erro e demais regras das skills/rules ativas estao respeitadas. Corrija desvios antes do resumo final.
  10. Antes de declarar a tarefa concluida ("pronto", "completo", "feito", "concluded", "done", etc.), inclua no resumo final a secao "Revisao de aderencia" com status ❌/✅ item-a-item da checklist objetiva. Se 2+ itens vermelhos, NAO use frase de conclusao — use formato WIP transparente. Build verde + tests verdes + lint verde NAO implica conclusao da tarefa quando ha refatoracao estrutural pendente.

Escopo dos Testes

  • Prefira testes unitários para regras de negócio, validações, cálculos e decisões.
  • Use testes de aplicação ou integração quando o risco estiver na orquestração, persistência, adapters, serialização ou contrato entre camadas.
  • Use testes ponta a ponta apenas quando o comportamento depender do fluxo completo.
  • Testes ponta a ponta sao OBRIGATORIOS quando o modulo persiste estado externo (arquivo, banco, fila, cache, broker, HTTP de saida) ou expoe entrypoint executavel (CLI, HTTP server, daemon, worker). Integration cobre o adapter contra o recurso; e2e cobre o fluxo completo cross-command/cross-request. Detalhes em testing-coverage-quality.
  • Não substitua um teste simples e próximo do comportamento por um teste amplo e lento sem necessidade.
  • Para CLI, teste parsing, saída, código de retorno e erros no adaptador de interface; testes em subprocesso são úteis, mas não devem ser a única cobertura da CLI quando a função de entrada pode ser chamada diretamente.
  • Ao iniciar um módulo novo, crie a estrutura mínima de testes junto da estrutura de produção.
  • A colocacao dos testes unitarios e definida pela rule da linguagem em .agents/rules: Go e Rust colocalizam; TypeScript e JavaScript preferem colocalizado; Python, Java, C# e PHP usam diretorio espelhado (tests/, src/test/java, projeto *.Tests). Siga a rule da linguagem em uso, nao um default unico.
  • Use tests/ no mesmo nivel de src/ para testes de integracao, contrato, ponta a ponta, fixtures compartilhadas e suites que cruzam modulos, mesmo quando os unitarios ficam colocalizados.
  • Quando testes ficarem dentro de tests/ (unitarios em linguagens como Python/PHP, ou integracao/contrato/e2e em qualquer linguagem), agrupe-os primeiro pelo TIPO de teste em subpastas como tests/unit/, tests/integration/, tests/contract/, tests/e2e/, tests/acceptance/. Dentro de cada subpasta de tipo, espelhe a estrutura de pastas do codigo alvo: um teste de src/<modulo>/<sub>/<arquivo> vira tests/<tipo>/<modulo>/<sub>/test_<arquivo> (ajustando o sufixo ao idioma do framework). Vale para qualquer tipo de teste, nao so unitario.
  • Excecoes por convencao da stack: em Java/Kotlin Maven/Gradle, a separacao de tipo e feita pelo sufixo do nome (*Test unitario, *IT integracao) sobre src/test/java espelhando src/main/java; em C#/.NET, a separacao e feita por projetos de teste distintos (MyApp.UnitTests, MyApp.IntegrationTests). Nesses casos, siga a convencao da stack em vez do agrupamento por subpasta de tipo.
  • Quando houver duvida sobre nomes e localizacao de testes, preserve o padrao local ja existente; na ausencia dele, siga a rule da linguagem.
  • Se a stack exigir outro layout, siga a convenção da stack e registre a razão na resposta final.

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. 9d ago First seen · 75 lines · 0 tokens per session scan A 9d39397f9552

Subscribe to this mod's changes

process-tdd is a skill published in the GitHub repository gianverdum/ai-dev-skillset (5 stars, last pushed 2mo ago), licensed MIT. It adds 175 tokens to every session and 1,640 once invoked, about $0.0009 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.