Home / muselinn / musepi · .musepi/skills/semantic-compression/SKILL.md · GitHub

semantic-compression skillA

semantic-compression is agent-read markdown (skill) from muselinn/musepi: Re-encode verbose prose into a dense telegraphic register — punctuation as connectives, label frames, verbless assertions — without losing normativity or precision. Use when compressing system prompts, tool/function descriptions, skill bodies, or agent instructions; reducing token count or context bloat; making documentation token-efficient for LLM input; or rewriting text in compressed notation..

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# Semantic Compression

Compression is **re-encoding, not word deletion**. Filtering function words out of an English sentence leaves a damaged English sentence (`System design: efficient process incoming data, multiple sources`). Instead re-frame each claim in a register whose grammar is punctuation and layout — then the function words have no work left and drop out on their own.

Target texts are load-bearing: tool descriptions, system prompts, skills. A model executes them cold, with no author present to disambiguate. Compression that forces a guess is a bug, not a saving.

## Procedure

0. **Density gate — check before touching anything.** Two signals, in order: (a) are articles and copulas already near-absent? (b) compress one representative section and measure the token delta. Already in this register (house-style prompt, tool doc, spec) or delta under ~10%? **STOP. Report that it is already dense and keep the original.** Bullet length alone is a weak signal — API literals and enumerations inflate it. Measured on a real house-style tool prompt: 853 → 778 tokens (8.8%), while that pass silently dropped a `NEVER assume …` rule, a throw condition, and a `full-res` detail. On…
…

Read the whole file at its exact version.

How to install

Latest version
mdr add muselinn/musepi/semantic-compression@git:20260823.eb76001
Exact content
mdr add muselinn/musepi/semantic-compression@sha256:11e5ee26fccf39d9

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_43fg4idptrcoevbm.svg)](https://markdownregistry.com/a/art_43fg4idptrcoevbm)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260823.eb76001 latest2026-08-23 eb76001 14,227 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (14227 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

muselinn/musepi · 17 stars · license MIT · pushed 2026-09-23 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_43fg4idptrcoevbm
GET https://markdownregistry.com/api/v1/resolve?ref=muselinn/musepi/semantic-compression
GET https://markdownregistry.com/api/v1/blob/11e5ee26fccf39d9eab3adcd02899acbb12ab2d2a4fb9a92220580bc34dab701

Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

More from muselinn/musepi

musepi-code-review skill
muselinn/musepi · .agents/skills/musepi-code-review/SKILL.md · 按 MusePi 仓库标准评审 PR/改动——AGENTS.md 代码质量规则(无 ReturnType、禁内联 import、prompt 进 .md)、GUI 规则(DialogFrame、hook…
git:20260916.ae02ea6 · audit A · 17 stars
musepi-find-simplifications skill
muselinn/musepi · .agents/skills/musepi-find-simplifications/SKILL.md · 在 MusePi 代码库找非显然的简化点——无生产消费者的公开方法/事件/配置项/helper、两个表示镜像同一事实、投机通用化、手搓重复中央工具或 Bun 内置的实现;产出有证据的候选清单(评审过、可进 PR +…
git:20260823.0df9512 · audit A · 17 stars
musepi-prose-standard skill
muselinn/musepi · .agents/skills/musepi-prose-standard/SKILL.md · MusePi 代码库的文案/注释标准——写"契约"而不是写过程:删推理过程、重复和装饰,一处事实一个家;覆盖 JSDoc、代码注释、prompt .md、i18n 文案、GUI…
git:20260909.1dfc4d6 · audit A · 17 stars
musepi-trim-cot-leakage skill
muselinn/musepi · .agents/skills/musepi-trim-cot-leakage/SKILL.md · 清理"思维链泄漏"类文案——死设计会话引用((决策 N)、审计码、草稿 §N、T4/P-I 阶段标签)、PR/栈视角("这个 PR 加了…")、变更叙述("以前/不再/旧版/这版")、评审对白("评审中否了…")、控制流叙述、含糊推迟("大…
git:20260823.0df9512 · audit A · 17 stars
system-prompts skill
muselinn/musepi · .musepi/skills/system-prompts/SKILL.md · Write system prompts, tool docs, and agent definitions. Project tag conventions + RFC 2119 keywords + dense…
git:20260916.c9f8599 · audit A · 17 stars
tool-prompt-optimization skill
muselinn/musepi · .musepi/skills/tool-prompt-optimization/SKILL.md · Optimize the description prompts an AI agent reads to learn its built-in tools (the `.md` files under prompts/tools/)…
git:20260823.eb76001 · audit A · 17 stars
AGENTS.md agents
muselinn/musepi · AGENTS.md
git:20260917.391022e · audit A · 17 stars
board-design skill
muselinn/musepi · packages/coding-agent/src/bundled-skills/board-design/SKILL.md · 桌面看板(board)的设计与读写规范——何时建板、怎么规划卡片布局、如何选 widget 类型并配 data/pos(避免类型数据错配和不可用布局)。触发:用户要求"看板/桌面卡片墙/把 X 放上看板/添加卡片"。
git:20260817.88c4ebc · audit A · 17 stars
musepi-contributing skill
muselinn/musepi · packages/coding-agent/src/bundled-skills/musepi-contributing/SKILL.md · MusePi 仓库提交规范助手——帮用户按项目模板与 CONTRIBUTING 规范起草/提交 GitHub issue、pull request、commit message 与 changelog,以及跑 release…
git:20260908.1c958e3 · audit A · 17 stars
musepi-extension-dev skill
muselinn/musepi · packages/coding-agent/src/bundled-skills/musepi-extension-dev/SKILL.md · 为 MusePi(TUI/CLI + 桌面 GUI)开发扩展——custom tools、extensions(生命周期/事件)、hooks、plugins、slash…
git:20260817.c5cdd0e · audit A · 17 stars
musepi-help skill
muselinn/musepi · packages/coding-agent/src/bundled-skills/musepi-help/SKILL.md · MusePi 产品使用帮助——回答"怎么用/是什么/出问题怎么办"类问题(TUI/CLI/桌面 GUI/看板/widget/配置/扩展/许可)。触发:任何关于 MusePi 本身的使用、功能、配置、故障问题。
git:20260909.1dfc4d6 · audit A · 17 stars
ui-ux-pro-max skill
muselinn/musepi · packages/coding-agent/src/bundled-skills/ui-ux-pro-max/SKILL.md · UI/UX design intelligence for web, mobile, and desktop. This skill should be used when designing, building, reviewing…
git:20260817.88c4ebc · audit A · 17 stars

Every file in muselinn/musepi

Other files named semantic-compression

semantic-compression skill
mlcyclops/lucidagentide · .agents/skills/semantic-compression/SKILL.md · Aggressively remove grammatical scaffolding LLMs reconstruct while preserving meaning-carrying content. Output may be…
git:20260618.7992a9c · audit A · 18 stars

Browse by kind, by grade A, or by owner.