# Send Clear Writing to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "clear-writing", "name": "Clear Writing", "source": "tencent", "type": "skill", "category": "效率提升", "sourceUrl": "https://clawhub.ai/wpank/clear-writing", "canonicalUrl": "https://clawhub.ai/wpank/clear-writing", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/clear-writing", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=clear-writing", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "README.md", "SKILL.md", "references/elements-of-style/01-introductory.md", "references/elements-of-style/02-elementary-rules-of-usage.md", "references/elements-of-style/03-elementary-principles-of-composition.md", "references/elements-of-style/04-a-few-matters-of-form.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "clear-writing", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-29T12:18:44.419Z", "expiresAt": "2026-05-06T12:18:44.419Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=clear-writing", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=clear-writing", "contentDisposition": "attachment; filename=\"clear-writing-0.1.0.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "clear-writing" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/clear-writing" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/clear-writing", "downloadUrl": "https://openagent3.xyz/downloads/clear-writing", "agentUrl": "https://openagent3.xyz/skills/clear-writing/agent", "manifestUrl": "https://openagent3.xyz/skills/clear-writing/agent.json", "briefUrl": "https://openagent3.xyz/skills/clear-writing/agent.md" } } ``` ## Documentation ### Clear Writing Write with clarity and force. This skill covers what to do (Strunk's rules), how to structure technical documentation (Divio patterns, templates), and what not to do (AI anti-patterns, doc anti-patterns). ### When to Use Use this skill whenever you write prose for humans: Documentation, README files, technical explanations API documentation, endpoint references, integration guides Tutorials, how-to guides, architecture docs Commit messages, pull request descriptions Error messages, UI copy, help text, comments Reports, summaries, or any explanation Editing existing prose to improve clarity If you're writing sentences for a human to read, use this skill. ### Limited Context Strategy When context is tight: Write your draft using judgment Dispatch a subagent with your draft and the relevant reference file Have the subagent copyedit and return the revision Loading a single reference (~1,000–4,500 tokens) instead of the full skill saves significant context. ### Elements of Style William Strunk Jr.'s The Elements of Style (1918) teaches you to write clearly and cut ruthlessly. ### Rules Elementary Rules of Usage (Grammar/Punctuation): Form possessive singular by adding 's Use comma after each term in series except last Enclose parenthetic expressions between commas Comma before conjunction introducing co-ordinate clause Don't join independent clauses by comma Don't break sentences in two Participial phrase at beginning refers to grammatical subject Elementary Principles of Composition: One paragraph per topic Begin paragraph with topic sentence Use active voice Put statements in positive form Use definite, specific, concrete language Omit needless words Avoid succession of loose sentences Express co-ordinate ideas in similar form Keep related words together Keep to one tense in summaries Place emphatic words at end of sentence ### Reference Files For complete explanations with examples: SectionFile~TokensGrammar, punctuation, comma rulesreferences/elements-of-style/02-elementary-rules-of-usage.md2,500Paragraph structure, active voice, concisionreferences/elements-of-style/03-elementary-principles-of-composition.md4,500Headings, quotations, formattingreferences/elements-of-style/04-a-few-matters-of-form.md1,000Word choice, common errorsreferences/elements-of-style/05-words-and-expressions-commonly-misused.md4,000 Most tasks need only 03-elementary-principles-of-composition.md — it covers active voice, positive form, concrete language, and omitting needless words. ### AI Writing Patterns to Avoid LLMs regress to statistical means, producing generic, puffy prose. Avoid: Puffery: pivotal, crucial, vital, testament, enduring legacy Empty "-ing" phrases: ensuring reliability, showcasing features, highlighting capabilities Promotional adjectives: groundbreaking, seamless, robust, cutting-edge Overused AI vocabulary: delve, leverage, multifaceted, foster, realm, tapestry Formatting overuse: excessive bullets, emoji decorations, bold on every other word Be specific, not grandiose. Say what it actually does. For comprehensive research on why these patterns occur, see references/signs-of-ai-writing.md. Wikipedia editors developed this guide to detect AI-generated submissions — their patterns are well-documented and field-tested. ### Document Types (Divio Framework) TypePurposeStructureREADMEFirst impression, project overviewTitle, description, quick start, install, usageTutorialLearning-oriented, guided experienceNumbered steps with expected outcomesHow-to GuideTask-oriented, solve a specific problemProblem statement → steps → resultReferenceInformation-oriented, complete and accurateAlphabetical or grouped, consistent formatExplanationUnderstanding-oriented, context and rationaleNarrative prose, diagrams, historyArchitecture DocSystem design, component relationshipsContext → components → data flow → decisionsAPI DocumentationEndpoint contracts, integration guideEndpoint → params → request → response → errors ### Inverted Pyramid Lead with the most important information. Each subsequent section adds detail. 1. What it does (one sentence) 2. How to use it (quick start) 3. Configuration options 4. Advanced usage 5. Internals / implementation details ### Problem-Solution 1. Problem — what goes wrong, symptoms, error messages 2. Cause — why it happens (brief) 3. Solution — step-by-step fix 4. Prevention — how to avoid it in the future ### Sequential Steps Every step is a single action with a verifiable outcome. 1. Step — one action, one verb Expected result: what the reader should see 2. Step — next action Expected result: confirmation of success ### Writing Rules RuleGuidelineExampleShort sentencesKeep under 25 words"The server restarts automatically after config changes."Active voiceSubject does the action"The function returns a promise" not "A promise is returned"Present tenseDescribe current behavior"This endpoint accepts JSON" not "will accept JSON"One idea per paragraphEach paragraph has one pointSplit compound paragraphs at the topic shiftDefine jargon on first useNever assume vocabulary"The ORM (Object-Relational Mapper) translates..."Second personAddress the reader directly"You can configure..." not "One can configure..."Consistent terminologyPick one term and stick with itDon't alternate between "repo" and "repository"Concrete over abstractSpecifics beat generalities"Returns a 404 status code" not "Returns an error" ### Code Examples in Documentation Every code example must follow these rules: Complete and runnable — copy-paste and execute without modification Annotated — comments on the non-obvious parts, not the obvious ones Progressive complexity — simplest case first, then advanced usage Language-tagged — always specify the language in fenced code blocks Current — examples must work with the documented version Minimal — show only what is relevant; strip unrelated boilerplate # Good: complete, annotated, minimal import httpx # Create a client with a base URL to avoid repeating it client = httpx.Client(base_url="https://api.example.com") # Fetch a user by ID — returns a User dict or raises for 4xx/5xx response = client.get("/users/42") response.raise_for_status() user = response.json() print(user["name"]) # "Ada Lovelace" ### README Template # Project Name One-line description of what this project does and who it is for. ## Quick Start The fastest path from zero to working. Three commands or fewer. ## Installation Prerequisites, system requirements, and step-by-step install. ## Usage Common use cases with code examples. Cover the 80% case. ## API Public API surface — functions, classes, CLI flags, endpoints. ## Configuration Environment variables, config files, and their defaults. ## Contributing How to set up the dev environment, run tests, and submit changes. ## License License name and link to the full LICENSE file. README rules: Keep the quick start under 60 seconds of reader time Include a badge row only if badges are kept current Link to deeper docs rather than bloating the README Update the README whenever the public interface changes ### API Documentation Pattern Document every endpoint with this structure: ### GET /users/:id Retrieve a single user by their unique identifier. **Authentication:** Bearer token required **Path Parameters:** | Parameter | Type | Required | Description | |-----------|--------|----------|----------------------| | id | string | Yes | The user's unique ID | **Response: 200 OK** {json response example} **Error Responses:** | Status | Code | Description | |--------|--------------|--------------------------| | 401 | UNAUTHORIZED | Missing or invalid token | | 404 | NOT_FOUND | User does not exist | Always document errors with: HTTP status, machine-readable error code, human-readable message, and resolution steps. ### Audience Adaptation AudienceContext LevelFocusToneBeginnerHigh — define terms, explain prerequisitesWhat and how, step by stepEncouraging, patientIntermediateMedium — assume basic knowledgeHow and best practicesDirect, practicalExpertLow — skip fundamentalsWhy, edge cases, tradeoffsConcise, precise Rules: State the assumed audience at the top of the document Link to prerequisite knowledge rather than re-explaining it Use expandable sections (
) for beginner context in expert docs Never mix audience levels in the same section ### Review Checklist Before publishing any documentation: Accurate — all code examples run, all commands work, all links resolve Complete — covers setup, happy path, error cases, and cleanup Consistent — terminology, formatting, and voice match the rest of the docs Readable — passes a cold read by someone unfamiliar with the project Scannable — headings, tables, and lists allow skimming for answers Examples work — every code block tested against the current version Links valid — no broken internal or external links Audience-appropriate — context level matches the stated audience Up to date — no references to deprecated features or old versions Spellchecked — no typos, no inconsistent capitalization ### Documentation Anti-Patterns Anti-PatternProblemFixWall of textReaders bounceBreak into sections with headings and listsOutdated docsErodes trustTie doc updates to PR checklists; date-stamp pagesNo examplesReaders can't apply abstract descriptionsAdd code examples for every public functionAssumed knowledgeExcludes beginnersDefine terms on first use, link to prerequisitesCopy-paste unfriendlyCode with $ prompts or line numbers breaks when pastedProvide clean, runnable code blocksScreenshot-only instructionsCan't be searched, go stale, inaccessiblePair screenshots with text and commands ### NEVER Do NEVER publish docs without testing every code example — broken examples destroy credibility faster than anything else NEVER write docs after the fact as an afterthought — write docs alongside the code; if you cannot explain it, the design needs work NEVER use "simply", "just", or "obviously" — these words shame readers who are struggling and add no information NEVER mix multiple audiences in one document — write separate beginner and advanced guides, or use clear section boundaries NEVER leave placeholder text in published docs — "TODO", "TBD", and "lorem ipsum" signal abandonment NEVER duplicate content across documents — link to a single source of truth; duplicates inevitably drift apart NEVER omit the date or version — readers must know if they are looking at current information NEVER use AI puffery words — pivotal, crucial, seamless, robust, groundbreaking, tapestry, and their ilk add nothing and signal lazy writing ## Trust - Source: tencent - Verification: Indexed source record - Publisher: wpank - Version: 0.1.0 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-04-29T12:18:44.419Z - Expires at: 2026-05-06T12:18:44.419Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/clear-writing) - [Send to Agent page](https://openagent3.xyz/skills/clear-writing/agent) - [JSON manifest](https://openagent3.xyz/skills/clear-writing/agent.json) - [Markdown brief](https://openagent3.xyz/skills/clear-writing/agent.md) - [Download page](https://openagent3.xyz/downloads/clear-writing)