docs-style

A documentation writing guide for creating, updating, or reviewing clear technical docs.

In plain words
What is it for?
Use it to structure tutorials, how-to guides, reference pages, and explanations with prerequisites, real examples, consistent terms, and expected results.
Why use it?
It helps prevent documentation from becoming vague, repetitive, hard to scan, or difficult to follow.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/continuedev/continue/docs-style
Any agent
npx skills add continuedev/continue --skill docs-style
Clone the repo
git clone --depth 1 https://github.com/continuedev/continue

Made for: Claude Code, Codex.

Per session 27 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,018 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00027 $0.01018
Opus 5 $0.00014 $0.00509
Sonnet 5 $0.00005 $0.00204
Haiku 4.5 $0.00003 $0.00102

Measured yesterday against content hash 130f0efbe0b2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

docs-style 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 yesterday.

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.

.claude/skills/docs-style/SKILL.md · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Docs Style Guide

Principles

  • Be concise — no filler. Make it easy to find what you're looking for
  • Task-oriented — frame around what the user is trying to do, not what the product can do
  • Progressive disclosure — guide from introduction to advanced use-cases. Don't throw users into the deep end
  • Real examples over abstract explanations — show, don't describe
  • Code snippets must be copy-pasteable — no placeholder values that silently break, no missing imports
  • Prerequisites up front — don't surprise the user halfway through
  • One topic per page — if you're covering two things, split it
  • Link, don't repeat — reference other docs instead of duplicating content
  • Scannable headings — skimming the TOC should reveal the page structure
  • Show expected output — after a step, tell the user what they should see
  • Consistent terminology — pick one term for a concept, use it everywhere
  • Screenshots/GIFs for key product features — use visuals when they teach faster than text
  • Know which type of doc you're writing — a tutorial (learning), a how-to (completing a task), a reference (looking something up), or an explanation (understanding why). Don't mix them in one page
  • Tutorials should be completable — a user following every step should end up with a working result, every time
  • Reference docs should be exhaustive and consistent — cover everything, use the same structure for every entry

Tone

  • Don't be patronizing — the reader is a developer. Don't tell them when to use something in a "when to use X vs Y" comparison table. If the distinction matters, state it plainly at the top of the relevant section in a sentence, then move on.
  • Respect the reader's time — open with the command or code, not a paragraph explaining what they're about to see. Lead with the thing, then explain.
  • No personality — the docs aren't a character. Don't try to be warm, clever, or endearing. No "Let's dive in!", no "The Magic of...", no "Pro Tip:", no emoji in headings. Developers see through it instantly and it reads like marketing copy wearing a docs costume. Just be direct and clinical. The docs serve information, they don't have a relationship with the reader.
  • Inline guidance over callout boxes — prefer weaving tips into the prose rather than using <Tip>, <Info>, <Warning>, etc. These components break reading flow and look heavy when overused. Reserve them for truly critical warnings (e.g. data loss, security). One per page is a good ceiling; zero is often fine.
  • Examples should feel real — use realistic file paths, realistic prompts, realistic tasks. Not > Tell me about the CLI but > @tests/auth.test.ts This test started failing after the last migration.
  • Examples earn their place — don't add "Example: Doing X" sections that are just English prompts in a code block. Examples are valuable when they demonstrate non-obvious syntax, flags, piping, or configuration. If the reader could figure it out from the rest of the page, skip the example.
  • No "Next Steps" sections — don't end pages with a "Next Steps" or "What's Next?" section with CardGroups linking to other pages. The sidebar navigation already does this. If a link to another page is relevant, put it inline where the context is, not in a generic footer.
  • Page title = sidebar title — the title in frontmatter should match the sidebar label. Drop sidebarTitle unless there's a genuine reason for them to differ. Don't stuff extra context into the page title (e.g., "Continue CLI (cn) Overview" → "Overview").
  • No subtitle/description in frontmatter — don't use the description field. The opening paragraph of the page should provide whatever context is needed. Metadata subtitles add clutter and duplicate what the prose already says.

Read the full file on GitHub · 42 lines

Changes

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.

  1. yesterday First seen · 42 lines · 27 tokens per session scan A 130f0efbe0b2

Subscribe to this mod's changes

docs-style is a skill published in the GitHub repository continuedev/continue (35,702 stars, last pushed 2d ago), licensed Apache-2.0. It adds 27 tokens to every session and 1,018 once invoked, about $0.0001 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 2026-08-30.

Related

Other skills, from other repositories

docs-architecture

Structure and write project documentation beyond the README. Use when docs have sprawled into an unnavigable pile, when users keep asking questions the docs already answer, when choosing a docs site generator (Docusaurus, MkDocs, VitePress, mdBook, Sphinx), or when planning a docs information architecture. Covers the…

the-open-agent/oss-skills · 116 tokens

bio-generative-design

Designs novel molecules using REINVENT 4 (de novo, scaffold decoration, linker design, R-group, molecular optimization), MolMIM, Diffusion-based generators (DiGress, DiffSMol), and JT-VAE with explicit handling of multi-parameter optimization (MPO), goal-directed scoring functions, transfer/reinforcement/curriculum…

PKU-YuanGroup/OpenAI4S · 119 tokens

agent-builder

Design and build AI agents for any domain. Use when users: (1) ask to "create an agent", "build an assistant", or "design an AI system" (2) want to understand agent architecture, agentic patterns, or autonomous AI (3) need help with capabilities, subagents, planning, or skill mechanisms (4) ask about Claude Code…

shareAI-lab/learn-claude-code · 125 tokens

copilotkit-self-update

Use when the user wants to update, refresh, or reinstall the CopilotKit agent SKILLS (the SKILL.md files that teach this agent about CopilotKit). NOT for updating the CopilotKit codebase or project — this is specifically about refreshing the skills/knowledge this agent has loaded. Triggers on "update copilotkit…

CopilotKit/CopilotKit · 144 tokens

edu-math-tutorial

数学题分步讲解视频的领域知识。适用场景:(1) 用户给出一道数学题并要求做讲解/ 解题视频;(2) 用户说"讲解这道题"、"生成解题视频"、"make a math tutorial"; (3) 需要中文数学教学视频,涉及方程、公式或几何图形。本 skill 只提供领域 知识(题目拆解、讲解节奏、旁白文案规范、版式与 KaTeX 排版风格);画面、 配音与成片一律用 Creator 原生 Element 与工具从源头构建。.

agentscope-ai/QwenPaw · 146 tokens

hive.browser-automation

Required before any hive-browser CLI command. The browser is driven from the terminal by running hive-browser ... --json via terminalexec — not via MCP tools. Teaches the browser lifecycle rules (the bridge attaches to the USER'S running Chrome — never kill or launch browser processes; timeouts are transport issues…

aden-hive/hive · 142 tokens