spec-generator

spec-generator is a skill for Claude Code, Codex from mouredev/hello-sdd. It costs 46 tokens per session (953 once invoked), scanned A, original, Apache-2.0.

A requirements-interview guide that turns a vague feature idea into an agreed Markdown specification. It records what will be built, what is excluded, and how to check each requirement.

In plain words
What is it for?
Use it to create, write, or review a feature specification. It reads existing project guidance and specifications, then creates a numbered spec file using the team’s template.
Why use it?
It reduces misunderstandings before implementation by asking about edge cases, errors, and scope. It also prevents missing details from being filled in with guesses.

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/mouredev/hello-sdd/spec-generator
Any agent
npx skills add mouredev/hello-sdd --skill spec-generator
Clone the repo
git clone --depth 1 https://github.com/mouredev/hello-sdd

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 spec-generator

README.md
[![agentmods](https://agentmods.dev/badge/skills/mouredev/hello-sdd/spec-generator.svg)](https://agentmods.dev/skills/mouredev/hello-sdd/spec-generator)
Your own site
<a href="https://agentmods.dev/skills/mouredev/hello-sdd/spec-generator"><img src="https://agentmods.dev/badge/skills/mouredev/hello-sdd/spec-generator.svg" alt="Measured on agentmods" height="20"></a>
Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 953 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 $0.00046 $0.00953
Opus 5 $0.00023 $0.00477
Sonnet 5 $0.00009 $0.00191
Haiku 4.5 $0.00005 $0.00095

Measured 4d ago against content hash ee3c562c6aa0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

spec-generator 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 4d 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.

habits-cli/.claude/skills/spec-generator/SKILL.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Generador de specs

Convierte una idea vaga en una especificación acordada. La spec es el contrato: si algo no está aquí, no se implementa.

Proceso

  1. Lee el contexto. docs/constitution.md si existe, y las specs previas de specs/ para respetar convenciones y no contradecir lo ya acordado.
  2. Entrevista al usuario. Preguntas de UNA en UNA, máximo 6, esperando respuesta antes de la siguiente. Céntrate en casos límite, comportamiento ante errores y qué queda fuera. No propongas soluciones técnicas: si el usuario pregunta "¿cómo lo harías?", redirige al QUÉ. Prioriza preguntas cuya respuesta cambie lo que hay que construir; descarta las que tengan una respuesta obvia por defecto.
  3. Elige el número. Mira specs/ y usa el siguiente libre con tres dígitos: specs/NNN-<nombre-en-kebab-case>/spec.md.
  4. Redacta usando spec-template.md de esta skill, sin saltarte secciones. Criterios de aceptación siempre en notación EARS, numerados como RF-1, RF-2, … Cada requisito debe ser verificable: si no se te ocurre cómo comprobarlo, está mal escrito.
  5. Marca lo que no sepas como [NECESITA ACLARACIÓN: pregunta concreta]. Nunca rellenes un hueco inventando: un hueco visible es información, una suposición silenciosa es deuda.
  6. Pide aprobación explícita al terminar. No pases al plan ni escribas código hasta tenerla.

Reglas

  • La spec describe QUÉ y POR QUÉ. Prohibido incluir stack, arquitectura, nombres de archivos, esquemas de datos, algoritmos o firmas de funciones: eso va en el plan.
  • Incluye siempre la sección "Fuera de alcance". Es la que evita que la funcionalidad crezca sola.
  • Un requisito, una frase. Si necesitas un "y" para unir dos comportamientos, son dos requisitos.
  • Sin adjetivos no medibles: "rápido", "intuitivo", "robusto" no son requisitos. Escribe el umbral o no lo escribas.
  • Idioma: el de la constitución del proyecto. Si no la hay, el del usuario.

Notación EARS

Read the full file on GitHub · 76 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 4d ago First seen · 76 lines · 46 tokens per session scan A ee3c562c6aa0

Subscribe to this mod's changes

spec-generator is a skill published in the GitHub repository mouredev/hello-sdd (437 stars, last pushed 7d ago), licensed Apache-2.0. It adds 46 tokens to every session and 953 once invoked, about $0.0002 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 2026-08-30.