knowledgebase skill
scriptsGenerates professional, multi-page HTML knowledge bases and technical study guides with sidebar navigation, Mermaid diagrams, and maintenance tracking.
- GitHub stars
- 21
- Files
- 7
- Last commit
- Nov 5, 2026
- License
- MIT
- Allowed tools
- —
What does the knowledgebase skill do?
knowledgebase generates professional, multi-page HTML knowledge bases and technical study guides with sidebar navigation, Mermaid diagrams, and maintenance tracking.
Summary generated from the skill's SKILL.md.
About the knowledgebase skill
Creating professional, multi-page HTML knowledge bases and technical study guides is what this skill enables for software engineers, data scientists, and architects. It produces self-contained, offline-capable websites that feature sidebar navigation, scrollspy table of contents, and light or dark theming. Instead of simple notes, it generates structured technical resources like team runbooks, service catalogs, and engineering field guides using a rigorous senior-engineer tone.
An agent should load this skill only when explicitly asked to build a specific deliverable such as a handbook or reference site. The process relies on a research-first pipeline where findings are stored in markdown files before being transformed into HTML using a specific page template. The output includes Mermaid diagrams for decision trees and pipelines, scenario drills for practitioner training, and maintenance pages designed to track volatile facts like pricing or service availability. It distinguishes itself by prioritizing decision frameworks and honest uncertainty over marketing prose, ensuring every technical claim is backed by verified research.
What an agent can do with it
- 1Build a team runbookGenerate a multi-page HTML site containing prioritized step sequences and decision trees for engineering processes.
- 2Maintain volatile technical documentationProduce a maintenance page with a master volatility table to track and verify changing technical facts.
Written from the skill's SKILL.md. Every use case is checked against a quote from it.
How to install the knowledgebase skill
Each command copies the skill folder into the client's skills directory. Restart the client or start a new session to load it.Install the knowledgebase skill in Plugin
/plugin marketplace add andisab/swe-marketplace /plugin install knowledgeware@swe-marketplace
Install the knowledgebase skill in Claude Code
~/.claude/skills/knowledgebase/SKILL.md
$ npx degit andisab/swe-marketplace/plugins/knowledgeware/skills/knowledgebase ~/.claude/skills/knowledgebaseInstall the knowledgebase skill in Claude Code (project)
.claude/skills/knowledgebase/SKILL.md
$ npx degit andisab/swe-marketplace/plugins/knowledgeware/skills/knowledgebase .claude/skills/knowledgebaseInstall the knowledgebase skill in Codex
~/.agents/skills/knowledgebase/SKILL.md
$ npx degit andisab/swe-marketplace/plugins/knowledgeware/skills/knowledgebase ~/.agents/skills/knowledgebaseInstall the knowledgebase skill in Gemini CLI
~/.gemini/skills/knowledgebase/SKILL.md
$ npx degit andisab/swe-marketplace/plugins/knowledgeware/skills/knowledgebase ~/.gemini/skills/knowledgebaseInstall the knowledgebase skill in Claude.ai
npx degit andisab/swe-marketplace/plugins/knowledgeware/skills/knowledgebase knowledgebase && zip -r knowledgebase.zip knowledgebase # claude.ai → Customize → Skills → + → Create skill → Upload a skill → knowledgebase.zip
Automated checks
Ran Nov 14, 2026Full report →Not verified yet. A skill needs a score of 80, no serious finding, and enough data for us to check. How checks work
- Source repository linked and reachableandisab/swe-marketplace/plugins/knowledgeware/skills/knowledgebase is public and reachablegithub.com/andisab/swe-marketplace/tree/HEAD/plugins/knowled ↗Pass
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit 2026-11-05github.com/andisab/swe-marketplace/commits ↗Pass
- No download-and-run or credential theft in scripts2 scripts read; no download-and-run, remote eval, base64 execution or credential uploadPass
- No hidden instructions in SKILL.mdNo override phrases, safeguard switches, credential requests or hidden charactersgithub.com/andisab/swe-marketplace/blob/HEAD/plugins/knowled ↗Pass
- Pre-approved tools are scopedNo pre-approved tools; the agent asks before using onegithub.com/andisab/swe-marketplace/blob/HEAD/plugins/knowled ↗Pass
- SKILL.md follows the Agent Skills formatdescription is 1027 characters; the limit is 1,024github.com/andisab/swe-marketplace/blob/HEAD/plugins/knowled ↗Finding
SKILL.md
View on GitHubKnowledgebase: Technical Knowledge-Base Generator
Generate a professional-grade, multi-page HTML knowledge base — the kind a senior architect keeps open in a browser tab and trusts in front of a customer. Not ordinary notes: a working reference with decision frameworks, verified numbers, honest uncertainty, and a maintenance story.
Operating persona and tone
Write as a seasoned senior solutions architect / staff engineer producing a reference for peers:
- Engineering methodology over business prose. Runbooks, decision trees, tables, and structured lists BEAT long-form copy. Context-setting is fine but brief.
- Clear, substantive, factual. No filler, no cheerleading, no "in today's fast-paced world."
- Decision-first ordering. Every page runs big decisions → medium decisions → fine detail. The top of a page is what a practitioner needs in the room; nuance lives below.
- Runbook-first openings. Where a page describes a process, open with a prioritized step sequence (numbered, imperative), then elaborate per step. Start with constraints (regulatory/compliance envelope first) when the domain involves platform or architecture selection.
- Every concept answers: what it is / why it exists / when to use / when NOT to use / trade-offs / common mistakes / production behavior.
- Honest uncertainty wins. Anything not confirmed against a primary source is marked
(unverified). An honest gap plus a verification plan beats a confident wrong number. Never invent figures. - Citations are clickable. Whenever a source has a URL, cite it as a live link (
<a href="…" target="_blank" rel="noopener">) — in refs callouts, footnotes, inline source mentions, and the maintenance page's volatility/fact-checking tables alike. A bare domain name or plain-text title the reader must google is a broken citation.
Process pipeline
Follow the phases in order. For a small resource (1–4 pages) run them solo; for larger sites, fan out subagents per the parallel-build playbook in references/build-process.md.
1. Scope and page map
Agree (or infer) the site's purpose, audience, and page list before writing anything. Organize navigation by the reader's workflow (phases of a process, decision order), not by source material. Define the full nav ONCE — every page carries identical nav markup. Plan a quick-reference page (numbers, traps, checklists) and a maintenance page (see §5) for any site with volatile facts.
2. Research first, write second
- Prioritize official documentation above everything; then reputable engineering sources. When sources disagree, say so and treat official docs as truth.
- Research findings land in a
research/folder as markdown, each file with source URLs, aLast verified: YYYY-MM-DDdate, and an explicit unverified-items section. Pages are built FROM research files, never from memory alone. - For multi-topic sites, run parallel research agents (one per domain) before any page writing. Give each a bounded brief and require the unverified-items section.
3. Build pages from the template
Use templates/page-template.html (this skill's directory). It provides: Anthropic-palette CSS with dark mode, adaptive width (clamp(900px, 86vw, 1400px)), sticky sidebar nav + mobile nav, scrollspy on-page TOC, horizontally scrollable table wrappers, and Mermaid wiring. Full mechanics, placeholder semantics, HTML idiom catalog (callouts, cards, scenario drills, stamps), and diagram rules: read references/formatting.md before writing the first page.
4. Verify before declaring done
Run the verification suite in references/build-process.md §Verification: zero leftover placeholders, every page's nav highlights itself, zero broken internal links, files end with </html>, no raw < inside Mermaid blocks, prev/next chain is a complete loop. A page is not done until it passes.
5. Design for decay
Volatile facts (prices, quotas, GA statuses, availability) rot. Every volatile table carries a Last verified stamp; any site with volatile facts gets a maintenance page built to the contract in references/maintenance.md: canonical section ids, a master volatility table whose rows are executable work orders (data class / complete page fan-out / authoritative source as a clickable link / change frequency / cadence enum / check method), the research-first update runbook, and a changelog that records no-change sweeps too. Treat >90-day-old stamps as stale. The contract is what lets the plugin's librarian agent (/knowledgeware:librarian sweep on a weekly schedule) maintain the site unattended — run the spec's §6 build-time checklist before declaring the site done.
6. Specs for decks
A deck is kept current through a spec page: a one-page site built to the maintenance spec's §7 (references/maintenance.md), with the deck's volatile facts as the master table and the usual maintenance sections. Build one when slideware asks for it beside a deck built from research, or when a user wants an existing deck maintainable: seed the master table from the deck's Sources: / Last verified: lines, mark guessed cadences for review, and follow the deck's style. Then slideware wires the deck to it: "spec" in an html deck's settings and data-teaches on its slides. A deck's plan (deck-plan.md, or <unit>.md beside the spec page) is not part of the spec; leave it where it is.
Brand override (brandware)
A brand or style named for the site ("in the AcmeCorp brand", "AAB style") resolves through the registry per the brandware skill's resolution contract (../brandware/SKILL.md §Resolution contract: the user's brands directory, then the plugin's brands, then the five generic styles; when nothing is named, the .knowledgeware and DEFAULT markers, telling the user what they resolved to). Map its palette and typography onto the template's :root CSS variables and the Mermaid themeVariables init per this skill's references/brand-mapping.md (the intent it implements: brandware's references/consumer-contract.md) — the template is fully variable-driven, so retheming touches only those two blocks. Chartware posters in a branded site follow the same brandbook (§chartware mapping — understated: neutral ink carries structure, accents annotate), and any data charts follow brandware's references/chart-styling.md. This skill's own fallback, when no brand is named or installed and no marker resolves: the template's default palette, and say so — brands are never a hard dependency.
Diagram policy (hybrid)
- Default: Mermaid, rendered by a locally vendored
mermaid.min.js(copy it into the site folder; the template wires a theme-aware init matched to the palette). Diagrams stay text-editable forever — right for a living document. - Posters: chartware (draw.io) for the 2–4 highest-value diagrams where visual communication is the point (the site's master flow, key decision gates). Export PNG into
diagrams/, keep the.drawioXML source beside it, embed with a caption noting the source file. Use thechartwareskill for authoring. - Mermaid rules:
<pre class="mermaid">blocks; ≤12 nodes per diagram; never raw<or>inside labels (write "under 200K" or use<);flowchart TDfor decisions,LRfor pipelines. - Genuine code samples, worked arithmetic, and config blocks stay as
<pre><code>— only diagrams get converted.
Content quality bar (per page)
- Dense
<section>blocks with h2/h3; many comparison tables (each wrapped in<div class="tscroll">); one or two Mermaid diagrams; callouts (tip= field tip,warn= trap,arch= architect's lens,refs= sources). - Scenario drills: 1–2 realistic practitioner situations per page in a dedicated end section, each linked to the section/workflow it exercises, with a
<details>expandable model answer. - Cross-link pages like chapters of a book; link official docs inline on load-bearing facts.
- 30–80KB per page is the typical depth band. Condense aggressively from research rather than padding.
- Decision tables carry a "choose on" rule — if a recommendation doesn't trace to a stated constraint or measured number, it's a preference; label it as one.
Anti-patterns (reject these in your own output)
Solution-first prose with no decision framework · long paragraphs where a table would do · unstamped volatile numbers · invented or "typical" figures · diagrams as ASCII art · pages organized by source material instead of reader workflow · marketing tone ("powerful", "seamless") · claiming completeness ("exhaustive") instead of scoping honestly ("80/20, deliberately not exhaustive").
Files in this skill
7knowledgebase skill questions
How do I install the knowledgebase skill in Claude Code?
Run npx degit andisab/swe-marketplace/plugins/knowledgeware/skills/knowledgebase ~/.claude/skills/knowledgebase. Claude Code loads skills from ~/.claude/skills (all projects) or .claude/skills (one project) and uses a skill when its description matches the task.
Which clients can use the knowledgebase skill?
Any client that reads the Agent Skills format: Claude Code, Codex, Gemini CLI and claude.ai (upload as a ZIP). Scripts in the skill run only where the client allows code execution.
What kind of documents can this skill generate?
It produces professional-grade HTML sites such as field guides, solution handbooks, service catalogs, team runbooks, onboarding documentation, and technical study resources.
How does the skill handle uncertain or unverified information?
The skill prioritizes honest uncertainty by marking any information not confirmed against a primary source as '(unverified)' rather than inventing figures.
What features are included in the generated HTML templates?
The templates include Anthropic-palette CSS with dark mode, sticky sidebar and mobile navigation, scrollspy on-page TOC, horizontally scrollable tables, and Mermaid diagram support.
How is the accuracy of volatile data maintained?
The skill includes a maintenance page with a master volatility table that tracks data changes, cadences, and authoritative sources to prevent information decay.
When should I use this skill?
You should explicitly invoke it by name (e.g., '/knowledgebase') or when asking to build a field guide, handbook, runbook site, or multi-page HTML study guide.