working-backwards

working-backwards is a skill for Claude Code from blake-simpson/belmont. It costs 53 tokens per session (5,412 once invoked), scanned A, original, Apache-2.0.

A structured method for writing a Working Backwards document, also called a PR/FAQ. It starts with the customer's needs and defines a product through a press release and frequently asked questions.

In plain words
What is it for?
Defining product vision, drafting a customer-facing announcement, answering likely questions, and recording the result in a PR/FAQ document.
Why use it?
It helps clarify the product's purpose and customer value before detailed planning or coding begins.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents; names the AskUserQuestion tool; mentions Codex.

Part of the belmont plugin — 17 skills shipped together

Good fit Defining product vision, drafting a customer-facing announcement, answering likely questions, and recording the result in a PR/FAQ document.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/blake-simpson/belmont/working-backwards
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.

Any agent
npx skills add blake-simpson/belmont --skill working-backwards
Clone the repo
git clone --depth 1 https://github.com/blake-simpson/belmont

Made for: Claude Code.

Or install belmont, the plugin that ships this one along with the rest of its 17 skills.

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 working-backwards

README.md
[![agentmods](https://agentmods.dev/badge/skills/blake-simpson/belmont/working-backwards.svg)](https://agentmods.dev/skills/blake-simpson/belmont/working-backwards)
Your own site
<a href="https://agentmods.dev/skills/blake-simpson/belmont/working-backwards"><img src="https://agentmods.dev/badge/skills/blake-simpson/belmont/working-backwards.svg" alt="Measured on agentmods" height="20"></a>
Per session 53 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,412 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 $0.00053 $0.05412
Opus 5 $0.00026 $0.02706
Sonnet 5 $0.00011 $0.01082
Haiku 4.5 $0.00005 $0.00541

Measured 7d ago against content hash c476070e31fb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

working-backwards 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 7d 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.

plugin/skills/working-backwards/SKILL.md · 450 lines

How it starts

The opening of the file, as written. The whole thing — 450 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Belmont: Working Backwards (PR/FAQ)

You are running an interactive Working Backwards session. Your goal is to work with the user to create a comprehensive PR/FAQ document — Amazon's methodology that starts with the customer and works backwards to define the product.

The output is a PR/FAQ: a one-page press release plus up to five pages of FAQs with an appendix. This format is used at Amazon, DAZN, and many other companies for product definition and decision-making.

This session requires ultrathink-level reasoning — deeply consider customer needs, market dynamics, and strategic implications before shaping the narrative.

CRITICAL RULES

  1. This is ONLY a strategic planning session. Do NOT implement anything.
  2. Do NOT create or edit any source code files.
  3. ONLY write to .belmont/PR_FAQ.md.
  4. Ask questions iteratively until the vision is clear.
  5. Always ask the user for clarification and approval before finalizing.

FORBIDDEN ACTIONS

  • Creating component files
  • Editing existing code
  • Running package manager or build commands
  • Making any code changes

Asking Questions (MANDATORY)

When you need to ask the user a question:

  1. Use your structured question tool (e.g. AskUserQuestion, or equivalent). This is NON-NEGOTIABLE when such a tool is available.
  2. Ask exactly ONE question at a time unless the tool itself is presenting a multi-select checklist. Wait for the user's answer before asking the next question.
  3. NEVER print the question as inline text AND use the tool. The tool call IS the question — do not duplicate it in your response body.
  4. NEVER ask questions as plain inline text when a structured question tool exists. No "Question 1: ..." followed by more text. Use the tool.
  5. Pick-list shape: questions should be keyboard-navigable numbered pick-lists. Put the recommended option first and label it (Recommended) when a recommendation is appropriate. Include a free-form "Type something" route and a "Chat about this" route when the structured tool supports those options.
  6. Single-select by default: use single-select pick-lists for normal planning decisions. Use multi-select only when the user genuinely needs to select more than one candidate; otherwise split the topic into sequential single-select questions.
  7. Codex fallback is strict: if you are running in Codex and the structured question tool is unavailable, do NOT approximate the pick-list in Markdown and do NOT continue the planning interview. Stop immediately and tell the user to restart the skill in Codex plan mode:
    • Product planning: /plan $belmont:product-plan <brief>
    • Technical planning: /plan $belmont:tech-plan <brief or feature>
  8. Non-Codex fallback: outside Codex, if no structured question tool exists at all, ask one clearly formatted plain-text question at a time and explicitly note that the structured picker is unavailable.

Read the full file on GitHub · 450 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. 7d ago First seen · 450 lines · 53 tokens per session scan A c476070e31fb

Subscribe to this mod's changes

working-backwards is a skill published in the GitHub repository blake-simpson/belmont (4 stars, last pushed 19d ago), licensed Apache-2.0. It adds 53 tokens to every session and 5,412 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 2026-08-31.