What is breaking-down-a-problem?

A structured workflow for turning a large or vague problem into a clear problem statement and three to seven smaller tasks.

Use it when scoping a project, investigating a stuck issue, or receiving a broad request such as fixing a system without knowing where to begin.

It makes difficult work easier to start by identifying the gap between what is happening and what should happen. It also selects the smallest useful task to begin with.

breaking-down-a-problem — Skill by redhuntlabs

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/redhuntlabs/wizard/breaking-down-a-problem

Any agent

npx skills add redhuntlabs/wizard --skill breaking-down-a-problem

Clone the repo

git clone --depth 1 https://github.com/redhuntlabs/wizard

Made for: Claude Code, Codex.

Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 999 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.

Model	Per session	Once invoked
Fable 5	$0.00028	$0.00999
Opus 5	$0.00014	$0.00500
Sonnet 5	$0.00006	$0.00200
Haiku 4.5	$0.00003	$0.00100

Measured 2d ago against content hash af7e462ddebc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

breaking-down-a-problem 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.

spells/thinking/breaking-down-a-problem/SKILL.md · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Breaking Down a Problem

What this does

Turns "this problem is too big" into a one-page problem statement plus a list of subproblems sized so you can start work on the smallest one today.

When to use

A problem feels too big or vague to begin
You've been stuck on something for a week with no progress
A new project is being scoped and you need to chunk it
Someone hands you "fix X" and you don't know where to start

What you bring (Inputs)

The problem in your own words (a sentence or paragraph)
The deadline or stakes (so we right-size the breakdown)

What you get (Output)

A precise 1-sentence problem statement
3-7 subproblems
The smallest one you can start on today

How it works (Steps)

This is a workflow.

Stages

Stage 1: Sharpen the problem statement

Re-write the problem as one sentence with three parts: observed (what is happening), desired (what should be happening), gap (the difference).

Example: "Onboarding takes new users 14 days to reach activation; we want under 3 days; current funnel drops 60% on day 2."

Stage 2: Identify the dimensions

What are the axes along which the problem could be split? Common axes: by user segment, by time/lifecycle, by component, by cause type, by responsibility, by cost.

Pick 1-2 axes that look most productive.

Stage 3: Decompose

Along the chosen axes, list 3-7 subproblems. Each subproblem should be:

Observable (you can tell whether it's solved)
Reasonably independent (solving one shouldn't require solving all)
Sized to be worked on (a day to a week, not a quarter)

Stage 4: Identify the smallest startable one

Of the subproblems, which can you start today with the inputs you have? That's your first move. Anything else is queued.

Stage 5: Validate

Read the breakdown back: does solving the subproblems together solve the original? If not, the decomposition missed something. Loop back.

Checkpoints

After Stage 1: problem statement has observed + desired + gap
After Stage 2: at least one axis chosen and rationale recorded
After Stage 3: 3-7 subproblems, each independent and sized
After Stage 4: one subproblem identified as today's starter
After Stage 5: subproblems collectively address the original gap

Read the full file on GitHub · 110 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.

2d ago First seen · 110 lines · 28 tokens per session scan A af7e462ddebc

Subscribe to this mod's changes

breaking-down-a-problem is a skill published in the GitHub repository redhuntlabs/wizard (9 stars, last pushed 4mo ago), licensed MIT. It adds 28 tokens to every session and 999 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.

Planning ai-agents ai-skills ai-tools ai-workflow automation claude-code

breaking-down-a-problem