aidenlx/zotlit / .agents/skills/docs-writing
docs-writing
VerifiedPlan and produce ZotLit's reader-facing docs: lessons at /learn, recipes, answer pages and reference pages at /docs, landing copy. Use when adding, rewriting, moving or retitling a docs page, writing a fact sheet or writer brief, or checking a draft's labels, anchors and search title.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 6.9 KB |
| references/brief/answer.opening.md | markdown | 358 B |
| references/brief/answer.shape.md | markdown | 1.8 KB |
| references/brief/lesson.opening.md | markdown | 348 B |
| references/brief/lesson.shape.md | markdown | 1.4 KB |
| references/brief/recipe.opening.md | markdown | 344 B |
| references/brief/recipe.shape.md | markdown | 2.0 KB |
| references/brief/shared.md | markdown | 3.2 KB |
| references/fact-sheet-model.md | markdown | 4.9 KB |
references/brief/shared.md
View on GitHub## Who you are writing for
A researcher: a PhD student, a postdoc, a lecturer. They read papers for a
living and they are good at it. They are an expert in their field and a
beginner in this software, and they know the difference.
They come to this lesson between two other tasks, late in the day, with a
chapter due. Reading is their work, so more reading is the last thing they
want. They will give this page a few minutes. If the first screen looks like
a manual, they close the tab and go back to copying page numbers by hand.
They already keep their papers in Zotero and they have highlighted some of
them. They have Obsidian because a colleague said it was good, or they
watched a video. Many have tried to connect the two before and stopped at a
template full of code. They are a little afraid of breaking their library.
What they want is small and concrete: their highlights next to their own
notes, a citation that goes into the draft, a Word file their supervisor can
open. They do not want to understand the software. They want to get back to
the chapter.
They do the lesson with Obsidian open beside the page. They read one step,
look away, do it, and look back. Each time they look back they need to find
their place at once and know whether the last step worked.
## The voice
Write as a colleague who uses this every day and has pulled a chair up to
their desk. You are showing, at their pace, on their screen. You say what to
click and what will appear, then you wait.
Use the plain words a researcher uses for their own work: paper, highlight,
note, draft, citation, reference list, supervisor. When the screen shows a
label, use the label exactly as the fact sheet gives it, so they can match
the page to the screen.
Be calm and exact. Treat the reader as an intelligent adult who is tired.
Confidence comes from each step working, so let the result carry it. When a
step can go wrong, say what they will see and what to do, in the same even
tone as the rest.
Where a step raises a worry about their library or their writing, answer it
in one sentence at that step: what changed, and what stayed as it was.
The course follows one story. The reader is writing Chapter 3 of a thesis,
"Presenting the results". Their Zotero collection is "Reproducibility". The
paper is Sandve et al. 2013, "Ten Simple Rules for Reproducible Computational
Research". Stay inside that story; a real paper and a real chapter make each
step easy to picture.
## Show first
The picture does the teaching; your words point at it. Each step already has
a short recording of the real app doing that step, and a still frame of the
result. For anything else the reader must know, choose the smallest view that
makes it clear, and write only what the view cannot say.
- A change: before and after, side by side.
- Where things live (Zotero, Obsidian, the reader's files): a small map of
labelled boxes and arrows.
- What an action changes and what it leaves alone: a two-column table.
- A choice between a few options: a table with one row per option.
You do not draw. For a view beyond the step's recording, write a figure
request: which view, what it shows, its labels, and a caption. A paragraph
that grows past three sentences means a view is missing; find the view.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |