Agent skill

Lildocs

by alloc in alloc/drizzle-plus

A skill your agent uses when authoring, reviewing, or restructuring Markdown documentation, especially docs architecture, page purpose, examples, technical-writing quality, and docs-change review.

MITAuto-check passedWriting & Content

Install Lildocs

skills CLI
$ npx skills add alloc/drizzle-plus --skill lildocs -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install alloc/drizzle-plus lildocs --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/alloc/drizzle-plus.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/lildocs .claude/skills/lildocs && rm -rf skills-src

Use ~/.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/

Facts

Skill name
lildocs
GitHub stars
174
Token cost
~3.1k tokens
SKILL.md length
1,446 words
Files
1
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when authoring, reviewing, or restructuring Markdown documentation, especially docs architecture, page purpose, examples, technical-writing quality, and docs-change review.

  • Works in 12 steps: Start with the reader's problem,… → Explain when the tool is useful and how… → Use plain, direct language; avoid… → …
  • Restructuring Markdown documentation
  • SKILL.md covers Read the Project, Page Purpose, General Technical… and Home Page, plus 8 more sections
  • Reaches aleclarson.github.io

What it does

Lildocs is an agent skill from alloc/drizzle-plus. Use when authoring, reviewing, or restructuring Markdown documentation, especially docs architecture, page purpose, examples, technical-writing quality, and docs-change review.

Its SKILL.md is about 3.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Writing & Content, covering Technical writing and Technical documentation. The repository describes itself as: A collection of useful utilities and extensions for Drizzle ORM. The licence is MIT.

When your agent uses it

  • Restructuring Markdown documentation
  • Especially docs architecture
  • Technical-writing quality
  • Docs-change review

Example prompts

  • “/lildocs”

Workflow steps

12 steps, taken from the first numbered list in SKILL.md.

  1. Start with the reader's problem, decision, or task.
  2. Explain when the tool is useful and how it differs from common alternatives.
  3. Use plain, direct language; avoid slogans, marketing, and unnecessary
  4. State limits, costs, prerequisites, and failure cases beside the benefits
  5. Support important claims with runnable examples, output, or observable
  6. Make examples complete, internally consistent, and safe to copy.
  7. Explain placeholders clearly and never present incomplete commands as
  8. Show how readers can verify both successful and failed operations.
  9. Organize content as a funnel: quick orientation, working example, mental
  10. Give each page one clear purpose and direct readers to the page that owns
  11. Keep advanced internals and optional workflows out of the required beginner
  12. Use consistent terminology across pages.

What it can do on your machine

Read from SKILL.md and the folder at commit 5840bdb. It shows what the files ask for, not the result of running them.

  • Tool permissions

    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.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown and jsonc).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • aleclarson.github.io

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Lildocs loads about 3.1k tokens when it runs. Until then it costs about 46 tokens; SKILL.md has 1,446 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~46
When it runs · the whole SKILL.md, loaded when a task matches
~3.1k

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.

Safety

Auto-check passed

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.

SKILL.md

The full file from alloc/drizzle-plus at commit 5840bdb, republished under its MIT licence (© alloc). 1,446 words, ~3,096 tokens.

Download SKILL.mdSave it as .claude/skills/lildocs/SKILL.md (or your agent's skills folder).
name
lildocs
description
Use when authoring, reviewing, or restructuring Markdown documentation, especially docs architecture, page purpose, examples, technical-writing quality, and docs-change review.

Technical Documentation

Help readers make correct decisions quickly: organize around tasks and concepts, state boundaries plainly, and prove claims with concrete examples.

Read the Project

Base documentation on the current project's facts, vocabulary, and support boundaries:

  • existing docs and README
  • docs root, folder layout, and generated navigation
  • package scripts and CI workflows
  • guides and fixtures used by the project
  • configuration, theme, font, and asset files

When publishing behavior matters, verify it from the installed package docs or source before writing about it:

text
node_modules/lildocs/docs/
node_modules/lildocs/

Use published docs only when local package docs are unavailable:

text
https://aleclarson.github.io/lildocs/

Treat generated-site constraints as content constraints: folder-based navigation, static links, headings and anchors, local search, diagrams, and assets all affect how readers find and trust the docs.

Page Purpose

Give each page one durable job. Before drafting, decide what the reader is trying to do, what they already know, what decision the page must support, and where they should go next.

General Technical Documentation Guidelines

Apply these guidelines to explanatory content throughout the docs. The home page's established slogan remains the intentional branding exception described below.

  1. Start with the reader's problem, decision, or task.
  2. Explain when the tool is useful and how it differs from common alternatives.
  3. Use plain, direct language; avoid slogans, marketing, and unnecessary jargon.
  4. State limits, costs, prerequisites, and failure cases beside the benefits they qualify.
  5. Support important claims with runnable examples, output, or observable results.
  6. Make examples complete, internally consistent, and safe to copy.
  7. Explain placeholders clearly and never present incomplete commands as pasteable.
  8. Show how readers can verify both successful and failed operations.
  9. Organize content as a funnel: quick orientation, working example, mental model, then reference.
  10. Give each page one clear purpose and direct readers to the page that owns deeper concepts.
  11. Keep advanced internals and optional workflows out of the required beginner path.
  12. Use consistent terminology across pages.
  13. Add diagrams only when relationships, ownership, flow, or branching are easier to understand visually.
  14. Validate commands, links, examples, formatting, and the generated documentation site before publishing.

Home Page

Treat the home page as the project's introduction, not as a map of the docs. Use the project's canonical display name as the H1 and its established slogan/tagline as the plain blockquote immediately below it. Prefer wording from the project's README or package metadata, and do not replace it with a generic title such as Introducing <name> or <name> Documentation.

md
# lildocs

> A lightweight CLI that turns Markdown docs into a static searchable
> documentation site.

This branding blockquote takes precedence over the general purpose-blockquote rule below; the rest of the home page can orient readers and link to the next steps.

For new pages in this project, follow the H1 with a purpose blockquote unless nearby docs use a different contract. The blockquote should clarify the page's real job: the decision it supports, the task it helps complete, or the boundary it draws.

md
# Command Line

> Build, preview, and publish flows start from different commands; this page
> keeps their flags and defaults separate so scripts stay small.

Weak purpose blocks restate the title or promise generic learning.

md
# Configuration

> Learn about configuration.

Prefer direct clarity over page-navigation language.

md
# Configuration

> Persistent site defaults belong in `config.json`; one-off choices belong in
> CLI flags for the current build or preview command.

Information Architecture

Organize docs by reader movement, not by source-code ownership. A useful docs set has a few clear shapes:

  • Overview: names the system, its moving parts, and the first meaningful decisions.
  • Task guides: complete a workflow from prerequisite to successful result.
  • Concept guides: explain boundaries, tradeoffs, and mental models.
  • Reference: supports lookup with complete, scannable facts.
  • Troubleshooting: starts from symptoms and leads to verification.

Put concepts at the point where readers need them. If a concept is needed by many pages, give it a canonical home and link to it instead of redefining it in each workflow.

Use file and folder names as navigation labels. Prefer short, stable nouns for concept areas and action-oriented names for workflows:

text
docs/
  index.md
  getting-started.md
  guides/
    publish.md
    customize-theme.md
  concepts/
    navigation.md
  troubleshooting.md

Keep prerequisite information before the steps that depend on it. Keep conceptual tradeoffs before the choice they influence. Put warnings immediately before the action they can change.

Mermaid Diagrams

Use a Mermaid diagram when visual structure helps the reader understand a relationship they would otherwise need to reconstruct from prose. Diagrams are especially desirable for:

  • workflows with meaningful branches, parallel steps, or feedback loops
  • lifecycle states and the transitions allowed between them
  • dependencies, ownership boundaries, or handoffs among several components
  • request or message sequences involving multiple actors
  • architecture overviews where grouping and connection are the point

Prefer flowchart for workflows, dependencies, and architecture; sequenceDiagram for interactions over time; and stateDiagram-v2 for lifecycle transitions. Keep each diagram focused on one idea, use short labels, and introduce it with prose that tells the reader what relationship to notice.

Do not add a diagram merely to decorate a page or restate a short linear list, definitions, headings, or a comparison that a table communicates more clearly. If layout, direction, grouping, or connection carries no additional meaning, use prose, a list, or a table instead.

Callouts

lildocs supports GitHub-style Markdown callouts. Use them when a detail changes how the reader should interpret or perform the surrounding task:

  • NOTE: useful context that prevents confusion but does not change the task
  • TIP: optional advice that improves the result or saves time
  • IMPORTANT: required information that readers must know before continuing
  • WARNING: risk, data loss, compatibility, or irreversible action to check
  • CAUTION: hazardous or easy-to-misuse behavior that needs extra restraint

Keep callouts close to the step, option, or concept they affect. Do not use a callout for ordinary prose, page summaries, or content that belongs in the main flow.

md
> [!NOTE]
> Search indexes are generated at build time, so changed pages require a new
> build before local search reflects them.

> [!WARNING]
> Delete the output directory only when it contains generated site files.
Show full SKILL.md (554 more words)Show less

Public API Documentation

When documentation touches TypeScript library APIs, keep factual API behavior near the source instead of creating hand-maintained docs/reference/ prose.

Default to this source-of-truth model:

  • public TSDoc owns symbol behavior, parameters, returns, errors, invariants, side effects, deprecations, and related APIs
  • docs/guides/ owns usage, composition, common workflows, and preferred defaults when those patterns belong in dedicated guide pages
  • concept docs own mental models, lifecycle, terminology, API-selection guidance, stable patterns, and anti-patterns
  • generated declarations own exact signatures and module shape

Treat the published surface as public:

  • package export-map entrypoints
  • source entry files intended for consumers
  • symbols reachable from generated declaration files
  • documented re-exports intended as API

Every public export should have at least a useful TSDoc summary. Add detailed tags when they clarify real behavior:

  • @param
  • @returns
  • @throws
  • @example
  • @remarks
  • @deprecated
  • @see

Do not document internal helpers as public API unless they are intentionally exported. If declarations expose internal-only symbols, prefer fixing the package boundary over documenting the leak as official API.

Writing Quality

Lead with the reader's next decision or action, then provide the smallest command, config, file tree, table, or Markdown pattern that completes it.

Prefer observable outcomes over vague benefits.

md
Weak:
This makes publishing easier.

Strong:
`pnpm run docs:build` writes static files to `./site` for CI to upload.

Keep terminology stable across pages. Change terms only when the distinction helps readers make a different decision.

md
Use `docs root` consistently.
Avoid switching between `source folder`, `content folder`, and `docs directory`
unless each term has a distinct meaning.

Use parallel structure when comparing options, fields, commands, or states. A table is often better than prose when readers need to scan for defaults, constraints, or differences.

md
| Option | Applies to | Default | Notes |
| --- | --- | --- | --- |
| `--out <dir>` | build, dev | `dist` | Directory for generated files. |

Qualify claims where the boundary matters. Prefer "when X, use Y" over broad rules that become false on the next page.

Example Discipline

Every non-trivial concept needs a nearby example. Non-trivial concepts include commands, config, file layout, Markdown syntax, workflow steps, API shapes, generated output, and errors.

Strong examples have four parts, even when some are only one sentence:

  • the situation that makes the example relevant
  • the smallest realistic input
  • the command, config, or content to use
  • the observable result or next check
md
Set a build output directory when CI expects artifacts in `./site`:

```bash
pnpm run docs:build -- --out ./site
```

After the command finishes, CI can upload `./site` as a static artifact.

For prose concepts, use before/after snippets rather than abstract advice.

md
Weak:
The build failed.

Strong:
The build failed because `docs/config.json` contains invalid JSON.

Example comments may explain intent, but they cannot carry information the surrounding prose omits. Use jsonc for JSON examples with comments so the comments are syntax highlighted correctly.

jsonc
{
  "navigation": {
    // Keep the getting-started page before generated folder entries.
    "order": ["getting-started.md", "guides/"]
  }
}

Reference Pages

Product reference pages should be complete inside their stated boundary and optimized for lookup speed. Put the boundary at the top, then use consistent tables, short subsections, and examples only where readers might choose incorrectly. For TypeScript API reference, prefer public TSDoc plus generated declarations over hand-maintained reference pages.

For commands, include syntax, required arguments, defaults, side effects, generated files, and failure cases that change user action.

For configuration, include field name, type, default, allowed values, merge or precedence behavior, and a minimal complete example.

For errors, start with the symptom, then list likely causes, verification steps, and the smallest fix that resolves each cause.

Review Checks

Before finishing docs changes, verify that:

  • each changed page has one clear reader job
  • navigation follows reader tasks rather than implementation trivia
  • duplicated explanations have a canonical home
  • claims are grounded in project files, tests, package docs, or source
  • every non-trivial concept has a nearby example
  • guide examples use project-realistic names, paths, and commands
  • links, headings, anchors, diagrams, and assets work in the generated site
  • terminology is consistent across changed pages

Run the project's available checks, such as formatting, linting, typechecking, tests, or a local docs build.

© alloc, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/lildocs of alloc/drizzle-plus.

Open the folder on GitHubat commit 5840bdb

Compare with similar skills

Lildocs 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.

Lildocs compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Lildocs this skillalloc/drizzle-plus174—~3.1kAutomated safety check: PassMIT
Beads Documentation Style Guidegastownhall/beads28k—~3.2kAutomated safety check: PassMIT
Technical Writing Standardcursor/plugins10k10 repos~2.4kAutomated safety check: PassNone
Heym Documentation Articlesheymrun/heym1.4k—~780Automated safety check: PassCustom licence
Developer Docs Technical Writervercel-labs/github-tools131—~3.9kAutomated safety check: PassMIT
Aholo Viewer Docsmanycoretech/aholo-viewer1.1k—~341Automated safety check: PassMIT

Similar skills

  • Sets the house style for the beads user docs: the canonical concept model, required terminology, prose and diagram conventions, and checks before docs work is done.

    28k GitHub stars~3.2k tokensUpdated today
    Writing & ContentAuto-check passed
  • Official

    Applies four layers of technical-writing rules to docs, RFCs, readmes, PR descriptions and commit messages so a tired engineer follows them on the first read.

    10k GitHub starsUsed in 10 repos~2.4k tokens
    Writing & ContentAuto-check passed
  • Creates and updates documentation articles for the Heym platform: category choice, manifest entry, markdown file and cross-links from existing pages.

    1.4k GitHub stars~780 tokensUpdated today
    Writing & ContentAuto-check passed
  • Developer Docs Technical Writer

    vercel-labs/github-tools

    Official

    Writes, reviews and edits developer documentation for SDKs, libraries and frameworks, from getting-started guides and API references to migration guides.

    131 GitHub stars~3.9k tokensUpdated 8 days ago
    Writing & ContentAuto-check passed
  • Aholo Viewer Docs

    manycoretech/aholo-viewer

    Guides writing and maintaining Aholo Viewer documentation: README, AGENTS.md, architecture notes, bilingual manual pages and AI collaboration guides.

    1.1k GitHub stars~341 tokensUpdated 13 days ago
    Writing & ContentAuto-check passed
  • Diataxis

    WebMCP-org/npm-packages

    Write technical documentation following the Diataxis framework by Daniele Procida.

    103 GitHub stars~1.7k tokensUpdated 3 days ago
    Writing & ContentAuto-check passed

Questions about Lildocs

What does Lildocs do?

A skill your agent uses when authoring, reviewing, or restructuring Markdown documentation, especially docs architecture, page purpose, examples, technical-writing quality, and docs-change review. Lildocs is an agent skill from alloc/drizzle-plus. Use when authoring, reviewing, or restructuring Markdown documentation, especially docs architecture, page purpose, examples, technical-writing quality, and docs-change review.

When should I use Lildocs?

Lildocs fits situations like: restructuring Markdown documentation; especially docs architecture; technical-writing quality; docs-change review.

How do I install Lildocs in Claude Code?

Run `npx skills add alloc/drizzle-plus --skill lildocs -a claude-code`. Or copy the skill folder (.agents/skills/lildocs in alloc/drizzle-plus) into .claude/skills/lildocs in your project. Claude Code loads it when a task matches its description.

How do I install Lildocs in Codex?

Run `npx skills add alloc/drizzle-plus --skill lildocs -a codex`. Or copy the skill folder (.agents/skills/lildocs in alloc/drizzle-plus) into .agents/skills/lildocs in your project. Codex loads it when a task matches its description.

Can I use Lildocs in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add alloc/drizzle-plus --skill lildocs -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/lildocs, .gemini/skills/lildocs, .github/skills/lildocs and .opencode/skills/lildocs in your project.

What does Lildocs need to run?

SKILL.md names no scripts, command-line tools or credentials: Lildocs is instructions for the agent only.

Does Lildocs access the network?

SKILL.md names 1 domain. In commands or code: aleclarson.github.io; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Lildocs safe to install?

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.

What licence does Lildocs use?

Lildocs is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Lildocs use?

About 3.1k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Lildocs?

Skills that share tags, products or a category with Lildocs: Beads Documentation Style Guide (gastownhall/beads, 28k stars), Technical Writing Standard (cursor/plugins, 10k stars), Heym Documentation Articles (heymrun/heym, 1.4k stars) and Developer Docs Technical Writer (vercel-labs/github-tools, 131 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Lildocs?

alloc (a GitHub organization) maintains it in alloc/drizzle-plus, which has 174 GitHub stars. The repository was last updated on August 22, 2026.

Source: alloc/drizzle-plus on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.