Plain Chinese

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites: none for the install itself: no account, no API key, no paid service, and nothing else to install first. The skill is two text files of writing rules only: no code, no script and no dependency to install. Note its output language before you start: the skill writes and rewrites in Simplified Chinese (see below). Install "plain-chinese" for me. It is a Simplified Chinese plain-writing playbook: 15 explicit rules, inspired by ASD-STE100 (simplified technical English) but not a Chinese version of it, that make Chinese explanations, steps and summaries readable in one pass — one idea per sentence, one meaning per term, active voice, no filler — while cutting no content at all: every fact, number, condition, exception and hedge stays. Repository: https://github.com/ChangWenC/understanding-ladder (the skill is the skills/plain-chinese/ folder at the pinned commit only; the repository also holds a second skill, understanding-ladder, which is a separate listing and is not installed here) Pinned version: install the code at commit 27b0957818f57a9c1d0ecdde1f50c2e3752f972c exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/ChangWenC/understanding-ladder into a temporary folder, check out commit 27b0957818f57a9c1d0ecdde1f50c2e3752f972c, and copy the skills/plain-chinese/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-19: [FINGERPRINTS] - SKILL.md: SHA-256 adab1f3c2bee90c1a9dadac13d183df3e26494ffae91de376f5cb0b8235ddc15 - references/examples.md: SHA-256 fcc3f9fbc119b660d8cc2a57d6a3aea3f36282a4f1f9f5c0110d327d9ca3142f If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full, plus references/examples.md, and tell me in plain words: what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The skill folder holds two files only: SKILL.md (5,493 bytes, the 15 rules, the rewrite method, the output format and the pre-delivery checklist) and references/examples.md (3,351 bytes, four worked before/after examples: a concept explanation, operating steps, explaining an agent's output, and a case of content that must not be cut). There is no code, no script and no dependency list in the folder: the skill is a set of writing rules for the agent itself. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. No file names an account, an API key, telemetry or a remote service, and the skill folder contains no URL at all. - Output language, stated plainly: the skill writes in Simplified Chinese. It is for writing Chinese technical explanations, concept notes, operating steps, study notes, reports and summaries, and for explaining what an AI or agent produced; its trigger phrases and its entire rule text are in Simplified Chinese, and the text it produces is Simplified Chinese. The only English in its method is an optional internal step for hard passages: draft a skeleton in English following the STE approach, then convert it to Chinese under the rules. If I want another output language, that is not what this skill does. - At use time, everything happens inside our conversation, on text I provide: the skill names no service, makes no network call, runs no command and installs nothing. Its 15 rules group into words (one term, one meaning, defined at first use and never swapped for a synonym; concrete words, no metaphors or jargon; direct verbs; numbers with units and clear range endpoints), sentences (one idea per sentence, about 25 characters at most; clear subject, active voice; affirmative sentences; condition before action; steps as numbered lists starting with a verb; pronouns that point to exactly one noun), paragraphs (one topic, five sentences at most, conclusion first; a list for three or more parallel items), what to delete (preamble, politeness and filler, empty adjectives, a closing paragraph that repeats the conclusion), and what to keep (every fact, number, condition, exception and hedge such as "usually" or "possibly"; code, commands, file paths, error text, quotations, proper nouns and formulas unchanged). - The skill has two strengths: regular (the default: all rules, sentence length as a target, reads as if written by a person — what it calls Karpathy's "80% of the way to ASD-STE100") and strict (only when I ask for it: sentence length as a hard limit, every term defined, no metaphors at all). By default it outputs only the rewritten text, with no preamble and no explanation of the rules; if a long sentence is kept so no content is lost, it adds one closing line naming what was kept and why, and only when I ask for a comparison does it output a table (original, rewrite, rule number). Before delivering, it runs a five-point self-check (one idea per sentence, one term per thing, clear subjects, no fact or hedge lost, no filler left). Its stated limit: it makes text understandable, not correct — a wrong source stays wrong, only easier to believe — so accuracy still needs a separate fact check. It is not for novels, marketing copy, speeches or text that lives by personal tone, which it would flatten. - The one "packages" signal in the folder is quoted example content, not an instruction of the skill: references/examples.md shows a before/after pair whose source paragraph mentions Node.js, npm install, npm run dev and editing PORT in a .env file, purely as a text to rewrite (the example teaches numbered steps and keeping commands unchanged). Do not run any of it and do not install anything because of it during this install or at use time. 3. When I say "yes, install", install it as a skill: copy the whole skills/plain-chinese folder into the agent's skills directory, in a folder named "plain-chinese", keeping both files, including the references/ subfolder. The upstream generic routes (npx skills add ChangWenC/understanding-ladder, or copying every skills/* folder at once) install the repository's two skills together; for this listing, copy only the skills/plain-chinese folder — the understanding-ladder folder, the examples/, docs/, plugin files and README files of the repository are not part of this skill and are not installed. 4. Verify with no network calls: frontmatter valid, both files in place. 5. Report what was installed, where, and what I still need to do myself (for example: provide the Chinese text or the topic myself; using the skill is a separate step that only starts when I ask for it, in Simplified Chinese, and needs no tool, account or key). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.