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 receptron/mulmoterminal --skill polishgit clone --depth 1 https://github.com/receptron/mulmoterminalWrote 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/receptron/mulmoterminal/polish)<a href="https://agentmods.dev/skills/receptron/mulmoterminal/polish"><img src="https://agentmods.dev/badge/skills/receptron/mulmoterminal/polish/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/receptron/mulmoterminal/polish"><img src="https://agentmods.dev/badge/skills/receptron/mulmoterminal/polish.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 | $0.00033 | $0.01309 |
| Opus 5.5 | $0.00013 | $0.00524 |
| Sonnet 5.5 | $0.00007 | $0.00262 |
| Haiku 4.5 | $0.00003 | $0.00131 |
Grade A, and why
blueprint-polish-polish 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 24d 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 — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Polish one file
Take the first file in .blueprint/polish.json whose status is "todo". Polish that file only; the
next round takes the next one. When the list is empty, the survey found nothing to polish: change nothing
and stop.
First, keep the original
Copy the file to .blueprint/originals/<the same path> before changing anything (create the folders).
The check compares against it, and it is how the person undoes a change.
What may change, and what may not
The document must still say the same thing. What changes is how it is said.
- May change: wording, sentence length and endings, spacing, spelling to the team's terms, splitting a long sentence or paragraph, removing padding.
- May not change — the check compares these with the original and fails on any difference:
- the headings, in order and word for word (other documents link to them);
- the articles, sections and items chaff reads as the document's structure (
chaff tree), so a reference such as "第3条第2項" or "Section 4.2" still points at the same place; - code blocks, byte for byte;
- link targets.
- Never add a fact, a number, a date or a name that is not in the original. Never drop one.
Follow STYLE.md when the answer scope includes the guide; otherwise, and when there is no scope, fix only what chaff reports.
Check it
sh <base pack>/checks/chaff.sh <the file> — with the same --genre as the survey when the answer kind
gives one (see <usecase pack>/kinds.json), and --show-baseline when the answer shelved asks for the shelved
findings too — must report no warning or error, except the ones you set aside.
Set a finding aside — never change the meaning to satisfy a rule — in one of two cases, by adding it to the
file's entry in .blueprint/polish.json:
{ "file": "guide.md", "before": 4, "status": "done",
"dismissed": [{ "rule": "max-sentence-length", "line": 12, "because": "meaning", "why": "条文の引用で、切ると原文と違ってしまう" }] }
"because": "wrong"— chaff misread the text (a heading it took for a sentence, a name it took for jargon). Each one becomes a draft report to chaff in the report step."because": "meaning"— chaff is right, but fixing it would change what the document says.
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.
- 24d ago Changed · +30 lines ae3560b9b063
- 25d ago Changed · +1 lines ee1b64ded8c2
- 26d ago Changed · +14 lines 97c6ae978628
- 27d ago First seen · 47 lines · 33 tokens per session scan A 766bf82b235a
blueprint-polish-polish is a skill published in the GitHub repository receptron/mulmoterminal (245 stars, last pushed yesterday), licensed MIT. It adds 33 tokens to every session and 1,309 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-28.
Other skills, from other repositories
moai-ref-ui-polish
UI polish and interface-completion reference: the small visual details — concentric border radius, optical alignment, shadow-vs-border, motion easing, typography smoothing, tabular numbers, icon stroke weight, hit areas — that separate polished interfaces from generic ones. Agent-extending skill that amplifies…
self-assessment
Interactive skill assessment with personalized learning path generation.
project-design
Roteador de processos de design e marca que carrega orientações para criar arquivos DESIGN.md completos para um projeto e suas aplicações de interface.
loom-swift
Swift language expertise for idiomatic, production-quality code.
loom-usage
Meta-orchestration skill for Claude driving loom itself.
loom-c
C language expertise (C11/C17/C23) for memory-safe, portable, production-quality systems code.