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.
npx agentmods add commands/robisson/build-like-amazon-agent-skills/wbgit clone --depth 1 https://github.com/robisson/build-like-amazon-agent-skillsWrote 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.
[](https://agentmods.dev/commands/robisson/build-like-amazon-agent-skills/wb)<a href="https://agentmods.dev/commands/robisson/build-like-amazon-agent-skills/wb"><img src="https://agentmods.dev/badge/commands/robisson/build-like-amazon-agent-skills/wb.svg" alt="Measured on agentmods" height="20"></a>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 5.1 | $0.00000 | $0.00539 |
| Opus 5 | $0.00000 | $0.00269 |
| Sonnet 5 | $0.00000 | $0.00108 |
| Haiku 4.5 | $0.00000 | $0.00054 |
Grade A, and why
wb 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Working Backwards — Start Here
Path resolution: All
skills/,agents/, andpatterns/paths in this command are relative to the plugin root directory. If not found in the working directory, resolve from the plugin installation path.
IMPORTANT: Each WB stage has a mandatory human approval gate. After completing each stage, present your work and ask the user to review before proceeding. Never skip stages or rush to PR/FAQ. The PR/FAQ is Stage 4 — it requires Stages 1-3 to be completed and individually approved first.
You are activating the Working Backwards meta-skill. This is Amazon's product discovery process: start from the customer and work backwards to the solution.
What to do
- Read the skill file at
skills/working-backwards/for full context. - Ask the user: "What product or feature are you thinking about?"
- Based on their answer, guide them through the five WB phases in order:
- Listen → Who is the customer? What do they need?
- Define → What is the specific problem we're solving?
- Invent → What is our proposed solution?
- Refine → Write the PR/FAQ document.
- Test → How will we validate and measure success?
- After each phase, STOP. Present the full output to the user. Ask them to review and approve before moving to the next phase. Do NOT proceed without explicit user confirmation.
- For each phase, load the corresponding skill from
skills/wb-*. - Produce artifacts in
docs/working-backwards/— customer profiles, problem statements, solution sketches, and the final PR/FAQ.
Approval Gates (Non-Negotiable)
After completing each stage, present your output and ask: "Here's what I have for [Stage]. Does this look right? Anything to adjust before we move to [Next Stage]?" Wait for explicit approval. If the user asks for changes, revise and re-present. Only proceed after they confirm.
Key Principles
- Customer obsession: every decision traces back to a customer need.
- Write the press release BEFORE writing code.
- If you can't write a clear FAQ answer, the idea isn't ready.
- Iterate: it's normal to loop between phases multiple times.
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.
- 6d ago First seen · 41 lines · 0 tokens per session scan A 1239d276fdde
wb is a command published in the GitHub repository robisson/build-like-amazon-agent-skills (14 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 539 tokens. 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.
Other commands, from other repositories
cancel-ralph
Cancel active Ralph Loop.
upgrade
Upgrade the skillshare CLI binary and/or the built-in skillshare skill.
release
Git geçmişinden hedef kitleye uygun sürüm notları oluştur.
spec-forge
Use when generating software specifications — full chain (Idea→Decompose→Tech Design + Feature Specs) or individual documents.
wp-set-attributes
The WP SET ATTRIBUTES command allows you to set the value of any attribute in a range, element, document. This command gives you access to any kind of 4D Write Pro internal attribute: character, paragraph, document, table, or image.
verify
Perform a non-destructive post-implementation verification gate validating the implementation against spec.md, plan.md, tasks.md, and constitution.md.