scaffold

An interactive guide for deciding whether a task needs an agent, command, skill, hook, or rule, then creating a matching template.

In plain words
What is it for?
Use it to describe what you want to build and answer a few questions about how it should start and what expertise it needs.
Why use it?
It removes the guesswork of choosing among Claude Code components when you want to automate something.

Command

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 commands/florianbruniaux/claude-code-plugins/scaffold
Clone the repo
git clone --depth 1 https://github.com/FlorianBruniaux/claude-code-plugins
Per session 51 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,361 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 $0.00051 $0.02361
Opus 5 $0.00026 $0.01180
Sonnet 5 $0.00010 $0.00472
Haiku 4.5 $0.00005 $0.00236

Measured 2d ago against content hash 6ca42c5b5887, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scaffold 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.

plugins/ai-methodology/commands/scaffold.md · 401 lines

How it starts

The opening of the file, as written. The whole thing — 401 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Code Scaffold Coach

Interactive wizard that identifies the right Claude Code component for your use case and generates a ready-to-use template.

Usage: /scaffold — no arguments needed. Start the conversation.


Phase 1 — Discovery

Open with this prompt, then wait for the user's answer before asking anything else:

What do you want to automate or build? One sentence is enough to start.

Once you have a rough idea, ask the following questions in order. Skip a question if a previous answer already answers it.

Q1 — Trigger

How is this triggered?

a) I run it myself with a command (e.g. /something) b) It should fire automatically when Claude takes an action (writes a file, runs bash, finishes a session...) c) It should apply all the time, every session, without me doing anything

  • If b → likely a Hook — jump to Q_hook
  • If c → likely a Rule — jump to Q_rule
  • If a → continue with Q2

Q2 — Domain expertise

Does this require deep, project-specific expertise? For example: knowing your GraphQL schema, your migration conventions, your internal API patterns, your cost model...

  • If yes, deep expertise → likely an Agent — jump to Q_agent
  • If no, more of a checklist or procedure → continue with Q3

Q3 — Complexity

How much context and logic does this involve?

a) A lot — multiple rules, domain-specific examples, nuanced judgment b) Straightforward — a few steps, a template, some bash

  • If a → likely a Skill
  • If b → likely a Command

Q4 — Reuse scope

Who needs this?

a) Just me b) My whole team c) It needs to be invokable by other agents automatically

  • c → strengthens Agent (needs a description: field that other agents can read)
  • b → strengthens Command or Skill (shared config)
  • a → can stay a simple personal Command

Q5 — Output type

What should happen at the end?

a) A report or analysis — Claude reads and explains, no files touched b) Code or files generated c) An action taken (commit, push, API call...) d) Claude's behavior changes permanently (always does X, never does Y)

Read the full file on GitHub · 401 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 · 401 lines · 51 tokens per session scan A 6ca42c5b5887

Subscribe to this mod's changes

scaffold is a command published in the GitHub repository FlorianBruniaux/claude-code-plugins (40 stars, last pushed 2mo ago), licensed MIT. It adds 51 tokens to every session and 2,361 once invoked, about $0.0003 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.