Sepia
Nanako0129/sepia
Make AI-generated writing read as human-written, in fiction and in professional prose.
Rules of good writing: coherence, clarity, structure, and a large anti-pattern catalog for catching machine-sounding text.
$ npx skills add alecs5am/ralphy --skill good-writing -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install alecs5am/ralphy good-writing --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/alecs5am/ralphy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/good-writing .claude/skills/good-writing && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "good-writing" agent skill from https://github.com/alecs5am/ralphy/tree/main/.agents/skills/good-writing into .claude/skills/good-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "good-writing", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/alecs5am/ralphy/tree/main/.agents/skills/good-writingType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add alecs5am/ralphy --skill good-writing -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install alecs5am/ralphy good-writing --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alecs5am/ralphy.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/good-writing .agents/skills/good-writing && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "good-writing" agent skill from https://github.com/alecs5am/ralphy/tree/main/.agents/skills/good-writing into .agents/skills/good-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "good-writing", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add alecs5am/ralphy --skill good-writing -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install alecs5am/ralphy good-writing --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alecs5am/ralphy.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/good-writing .cursor/skills/good-writing && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "good-writing" agent skill from https://github.com/alecs5am/ralphy/tree/main/.agents/skills/good-writing into .cursor/skills/good-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "good-writing", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/alecs5am/ralphy.git --path .agents/skills/good-writing--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add alecs5am/ralphy --skill good-writing -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install alecs5am/ralphy good-writing --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alecs5am/ralphy.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/good-writing .gemini/skills/good-writing && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "good-writing" agent skill from https://github.com/alecs5am/ralphy/tree/main/.agents/skills/good-writing into .gemini/skills/good-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "good-writing", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install alecs5am/ralphy good-writingInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add alecs5am/ralphy --skill good-writing -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/alecs5am/ralphy.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/good-writing .github/skills/good-writing && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "good-writing" agent skill from https://github.com/alecs5am/ralphy/tree/main/.agents/skills/good-writing into .github/skills/good-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "good-writing", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add alecs5am/ralphy --skill good-writing -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install alecs5am/ralphy good-writing --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alecs5am/ralphy.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/good-writing .opencode/skills/good-writing && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "good-writing" agent skill from https://github.com/alecs5am/ralphy/tree/main/.agents/skills/good-writing into .opencode/skills/good-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "good-writing", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
good-writingRules of good writing: coherence, clarity, structure, and a large anti-pattern catalog for catching machine-sounding text.
Good Writing is an agent skill from alecs5am/ralphy. Rules of good writing: coherence, clarity, structure, and a large anti-pattern catalog for catching machine-sounding text. A reference skill loaded from other skills whenever prose is WRITTEN or EDITED — video narration scripts, captions, seo-articles, social copy, unit descriptions, docs. The core idea: text must explain the way you would explain to a colleague over coffee — not like a textbook, not like marketing, not like an academic paper. Supplies the read-it-aloud test, the coherence rule (every sentence…
Its SKILL.md is about 2.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/antipatterns.md`).
It sits in Writing & Content, covering Blog and article writing, Text to speech and voice and Humanizing AI text. The repository describes itself as: Open-source desktop app for content creation, with an agent runtime and standalone CLI. The licence is Apache-2.0.
Read from SKILL.md and the folder at commit 8d139f0. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Links to these hosts (documentation or services it may open):
github.comFrom URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Good Writing loads about 2.1k tokens when it runs, and up to ~4.6k if it reads all its reference files. Until then it costs about 244 tokens; SKILL.md has 1,105 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from alecs5am/ralphy at commit 8d139f0, republished under its Apache-2.0 licence (© alecs5am). 1,105 words, ~2,147 tokens.
.claude/skills/good-writing/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.Ported (translated + condensed) from the Russian-language good-writing skill in
mikeshu2001/content-factory.
The original is the source of truth for Russian-specific grammar rules; this port
keeps every transferable rule in English. Attribution: mikeshu2001. The original
files are not vendored verbatim because this repo is English-only on disk
(docs/developing-ralphy.md).
This is not a list of bans to tick off. It is a toolset for moments of doubt. The goal: text should explain the way a person explains to a colleague over coffee. Not like a manual, not like marketing, not like an academic paper.
Before editing anything, ask:
If a construction is allowed by the rules but sounds clumsy — rewrite it. If it is banned but sounds natural and precise — keep it. Rules are for doubt, not a final judge.
Calm, informational. No marketing, no exclamation, no attempts to impress. Talk WITH the reader, not AT them. Direct address is fine; chumminess is not. Be direct: if something does not work, say so plainly; if something matters, say it first, not after a wind-up.
Every next sentence answers the question that arises after the previous one. The reader must never guess how two thoughts connect.
Connections must be semantic, not verbal. If one sentence logically continues another, do not add "because / therefore / also" — those words are only for places where the link is not obvious without them. Paragraphs must also continue each other; a paragraph change must not feel like a topic jump.
Progressive disclosure. A new concept appears only after the reader has developed a need for it. Do not open with "X is Y". First show the situation, limitation, or question that naturally leads to X — then the term is an answer, not a dictionary entry. After drafting, strip the headings and read the text as one flow: if the paragraph order is unclear without headings, the logic (not the headings) is broken.
Quantitative steps. Do not skip intermediate steps in a process. If a process has stages with counts ("one video = 5 scenes, one poster per scene"), state them explicitly and show how each stage's output feeds the next.
Paragraphs are 2-4 sentences, one thought each. Exception: explaining a new concept or a process may take 4-6 sentences — do not shred it into fragments.
Sentences are complete, with a subject and a verb. Every sentence must move the thought forward: if you can delete it and the paragraph's meaning survives, delete it. One block = one job (explain, show an example, instruct, warn) — do not mix jobs inside a block.
No "In today's world..." style intros, no sentences that announce information is coming. If the block heading is "Why the tool fails", the reader knows why they are here — answer the heading immediately. Do not write transition paragraphs between blocks; the heading already gives context.
Explain professional terms on first use, or replace them with plain descriptions. Test: "would someone outside this field understand?" IT jargon ("out of the box", "hardcode", "natively") gets rewritten for a general audience.
Headings are navigation, not decoration: name the action or the concrete subject, from the reader's perspective ("Which agents to try", not "Products on the market"). One heading = one question. No clickbait add-ons.
Lists only where they belong: enumerations, checklists, step sequences. In normal narration, write prose. Bold sparingly — key terms on first mention and real warnings. Code blocks for anything the reader may want to copy (prompts, requests, templates). Number only sequential steps; unordered points get bold lead-ins, not numbers.
Use minimally. A dash often hides a skipped thought and makes text choppy.
Prefer a verb: "The tool scores calls quickly", not "Call scoring tool — fast
and accurate". (In this repo dashes in VO scripts also produce long ElevenLabs
pauses — memory elevenlabs-multilingual-v2-em-dash-pacing.)
The full catalog — neuromarkers, hidden neuromarkers, syntactic weight, date
pinning, service-description bloat, marketing tone, empty generalities,
duplicate blocks, and the final check-question list — lives in
references/antipatterns.md. Read it before the
final pass on any text.
Pair with /humanizer (Wikipedia
"Signs of AI writing" catalog) — humanizer catches English-specific AI tells;
this skill owns structure, coherence, and tone.
The original skill carries rules that only make sense in Russian grammar. When writing Russian copy, consult the original source files for: verbal nouns used instead of verbs (nouns ending in -tsiya/-nie/-ka paired with adjectives), short passive participles as predicates (forms like "sobrano/zadan"), bookish copulas ("yavlyaetsya", "predstavlyaet soboy"), bureaucratic adverbials ("po mere", "v ramkakh", "v khode"), aspect/tense agreement inside a paragraph, and the ban on the letter "yo". Source: https://github.com/mikeshu2001/content-factory/blob/main/skills/good-writing/antipatterns.md
© alecs5am, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 1 other file (references) in .agents/skills/good-writing of alecs5am/ralphy.
Open the folder on GitHubat commit 8d139f0
Good Writing next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Good Writing this skillalecs5am/ralphy | 138 | — | ~2.1k | Automated safety check: Pass | Apache-2.0 | |
| SepiaNanako0129/sepia | 3.1k | — | ~3.6k | Automated safety check: Pass | MIT | |
| Humanizer Zhai-zixun/humanizer-zh | 179 | — | ~1.2k | Automated safety check: Pass | MIT | |
| Content Strategy And Assemblyjacob-dietle/context-os | 111 | — | ~3.1k | Automated safety check: Pass | MIT | |
| Linkedin Writerflaqai/backlink_skills | 756 | — | ~6k | Automated safety check: Pass | MIT | |
| Writewaynesutton/markdown-site | 627 | — | ~3k | Automated safety check: Pass | MIT |
Nanako0129/sepia
Make AI-generated writing read as human-written, in fiction and in professional prose.
ai-zixun/humanizer-zh
Remove signs of AI-generated, translated, or overly mechanical Chinese prose.
jacob-dietle/context-os
This skill should be used when producing content (newsletter posts, blog posts, LinkedIn posts) from existing corpus material.
flaqai/backlink_skills
LinkedIn-native long-form article and newsletter writing workflow for LinkedIn and Google-to-LinkedIn topic discovery, business-depth research, professional thought leadership, evidence-led…
waynesutton/markdown-site
Writing style guide for technical content, social media, blog posts, READMEs, git commits, and developer documentation.
sickn33/agentic-awesome-skills
Generate clean, human-sounding, SEO-optimized WordPress blog posts with optional Yoast metadata, JSON-LD schema markup, and image SEO planning.
alecs5am/ralphy
GSAP animation reference for HyperFrames. An agent skill from alecs5am/ralphy.
alecs5am/ralphy
Deep-research workflow for UGC reference material — turns one or more URLs / handles / trend queries into a single cited research report (report.md + sources.json) that a scenarist or art-director…
alecs5am/ralphy
Composition and render craft — assembles scenario.json plus asset-manifest.json into a HyperFrames HTML composition and renders the mp4.
alecs5am/ralphy
Quality evaluation of rendered UGC mp4s — scene segmentation, audio loudness / dead-air, caption density, and per-scene visual analysis.
alecs5am/ralphy
End-to-end orchestration — the wrapper that drives the whole production contract across roles, plus batch production.
alecs5am/ralphy
Scenario and script craft — writes and reworks the scene-by-scene scenario.json: hook, beat structure, per-scene VO, on-screen text, pacing, and the language/aspect pre-flight.
Categories
Rules of good writing: coherence, clarity, structure, and a large anti-pattern catalog for catching machine-sounding text. Good Writing is an agent skill from alecs5am/ralphy. Rules of good writing: coherence, clarity, structure, and a large anti-pattern catalog for catching machine-sounding text.
Good Writing fits situations like: editing any reader-facing prose: a VO script; an article draft; the user asks to check / rewrite a text in plain human language (in any language they phrase it); pairs this skill with /humanizer for an anti-AI pass.
Run `npx skills add alecs5am/ralphy --skill good-writing -a claude-code`. Or copy the skill folder (.agents/skills/good-writing in alecs5am/ralphy) into .claude/skills/good-writing in your project. Claude Code loads it when a task matches its description.
Run `npx skills add alecs5am/ralphy --skill good-writing -a codex`. Or copy the skill folder (.agents/skills/good-writing in alecs5am/ralphy) into .agents/skills/good-writing in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add alecs5am/ralphy --skill good-writing -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/good-writing, .gemini/skills/good-writing, .github/skills/good-writing and .opencode/skills/good-writing in your project.
SKILL.md names no scripts, command-line tools or credentials: Good Writing is instructions for the agent only.
SKILL.md names 1 domain. As links in the text: github.com. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Good Writing is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 2.1k tokens (SKILL.md is roughly 8.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2.5k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Good Writing: Sepia (Nanako0129/sepia, 3.1k stars), Humanizer Zh (ai-zixun/humanizer-zh, 179 stars), Content Strategy And Assembly (jacob-dietle/context-os, 111 stars) and Linkedin Writer (flaqai/backlink_skills, 756 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
alecs5am (a GitHub user) maintains it in alecs5am/ralphy, which has 138 GitHub stars. The repository holds 28 skills in this directory. The repository was last updated on September 22, 2026.
Source: alecs5am/ralphy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.