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 skills add MrBlushu/blushu-design-skills --skill before-we-make-a-messgit clone --depth 1 https://github.com/MrBlushu/blushu-design-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/skills/mrblushu/blushu-design-skills/before-we-make-a-mess)<a href="https://agentmods.dev/skills/mrblushu/blushu-design-skills/before-we-make-a-mess"><img src="https://agentmods.dev/badge/skills/mrblushu/blushu-design-skills/before-we-make-a-mess/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mrblushu/blushu-design-skills/before-we-make-a-mess"><img src="https://agentmods.dev/badge/skills/mrblushu/blushu-design-skills/before-we-make-a-mess.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00121 | $0.01639 |
| Opus 5 | $0.00060 | $0.00820 |
| Sonnet 5 | $0.00024 | $0.00328 |
| Haiku 4.5 | $0.00012 | $0.00164 |
Grade A, and why
before-we-make-a-mess 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 11d 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Before We Make a Mess
Operating contract
Make an uncertain product decision explicit and evidence-aware before consolidating a UX/UI solution. Produce a decision, a focused learning action, or a bounded handoff—not a delivery roadmap or an encyclopedic checklist.
Never invent users, evidence, metrics, research findings, provenance, or confidence. Keep facts/evidence, inferences, assumptions, and unknowns visibly distinct.
If the request is solely about interaction patterns, interface usability, visible text, project delivery, or technical implementation, route it to the relevant specialist and stop. When discovery and design are mixed, resolve the blocking product decision first and hand off only the remaining responsibility.
Workflow
- Define the decision. Rewrite the request as a concrete commitment to accept, defer, reframe, or reject. Name whether it concerns an opportunity, candidate solution, learning activity, or readiness gate.
- Frame the opportunity. State target, situation, problem, current alternative, desired outcome, success signal, urgency, and material constraints. Separate the outcome from the proposed solution. Mark missing information rather than filling it plausibly.
- Build the claim ledger. Classify each decisive claim as
evidence,inference,assumption, orunknown. Preserve provenance, limitations, and contradictions. Treat stakeholder opinions and model output as claims, not evidence. - Prioritize risks. Translate assumptions into consequences if false. Consider problem/value, use/comprehension, feasibility, business/context, and harm only when they can change the decision. Keep at most five active risks by default.
- Choose the next learning action. Target the highest-priority risk with the smallest credible research activity, experiment, prototype, spike, or bounded live test. Predeclare observable signals and the decision each result would change.
- Take a position. Select
proceed,continue discovery,reframe/pause, orstop. Explain the evidence used, residual risk, reversibility, and what the state authorizes now. Do not presentproceedas proof orstopas permanent truth. - Define the handoff. Pass only the problem frame, decisive claims, accepted risks, constraints, prioritized principles, open questions, and review triggers. Do not perform the recipient skill’s work.
What ships with it
5 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.
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.
- 11d ago First seen · 88 lines · 121 tokens per session scan A a641482e775a
before-we-make-a-mess is a skill published in the GitHub repository MrBlushu/blushu-design-skills (4 stars, last pushed 1mo ago), licensed MIT. It adds 121 tokens to every session and 1,639 once invoked, about $0.0006 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.
Other skills, from other repositories
anti-ui-slop
Stop coding agents from shipping generic UI. Extend the product's design system, use UIZZE evidence only when useful, cover required states, and inspect the rendered result.
image-director
The Creative Design Manager — manages ALL visual design production under 1ai-content. Delegates to logo design, icon design, banner design, UI/UX, and brand design system. Has memory via 1ai-hub brain, follows 1ai-rules, and provides visual assets to Content Director, Marketing Director, and Sales Director. Use when…
tsq-ui
A method for planning screens and user journeys before building them. It uses simple wireframes—rough screen layouts—and flow diagrams showing how users move through an interface.
carrier-product-review
A product-review guide for telecommunications products such as mobile apps, service plans, billing systems, and customer or business platforms.
frontend-design
Design and implement context-fit frontend interfaces with deliberate art direction, accessible interaction, responsive behavior, complete states, and rendered verification. Use when creating or substantially reworking pages, components, product workspaces, dashboards, marketing sites, editorial surfaces, commerce…
design-system
Document and define UI design system tokens, components, and guidelines.