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.
npx agentmods add skills/quantumentangled-dev/ai-docs-plugin/createnpx skills add quantumentangled-dev/ai-docs-plugin --skill creategit clone --depth 1 https://github.com/quantumentangled-dev/ai-docs-pluginWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00205 | $0.01521 |
| Opus 5 | $0.00102 | $0.00760 |
| Sonnet 5 | $0.00041 | $0.00304 |
| Haiku 4.5 | $0.00020 | $0.00152 |
Grade C, and why
create scanned grade C with 1 finding 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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
**Hosting-neutral:** `create` doesn't publish. Visibility = `publish` skill's concern. Sensitive content → ask about publishing intent, don't refuse outright. How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
create
Author or edit a paired .md + .html in CWD. Dispatches ai-docs-builder; main thread doesn't write files or run build scripts.
When to fire
Should trigger — creation: "create an HTML deck about ...", "build a presentation on ...", "draft a report on ...", "/ai-docs:create some-topic", etc.
Should trigger — iteration: "edit the login diagram ...", "update the rollout in auth-migration ...", "modify slide 3 of q3-roadmap deck", "remove the appendix from bar.md", "fix the typo in q2-metrics deck", etc.
The orchestrator recognizes iteration when the user references a doc type (deck, report, page) or a slug-like name matching an existing .md/.html in CWD. Pass the existing path in BRIEF; the builder's iteration workflow takes over.
Should NOT trigger:
- "publish my deck" →
publishskill - "summarize this URL" →
readskill - "make a diagram" (no doc context) →
mermaidskill
Dual-purpose
AI consumers read .md; humans view .html. Use semantic tags, no JS-rendered DOM, sane heading outlines.
Static-output constraints
Plain HTML + markdown. No runtime, no server, no API endpoints, no databases (inline data or fetch from view-time-reachable URL). Server-side logic needed → surface constraint + offer (a) static snapshot or (b) different tool.
Hosting-neutral: create doesn't publish. Visibility = publish skill's concern. Sensitive content → ask about publishing intent, don't refuse outright.
Output convention
./<slug>.md— AI source (always)./<slug>.html— rendered, with<link rel="alternate" type="text/markdown" href="<slug>.md">in<head>
Slug = short kebab-case derived from brief.
Theme picker (creation only)
Skip entirely on iteration — the builder scrapes the active theme from the existing .html marker and reuses it. Skip also if BRIEF already names a theme slug (e.g. "create a forest-canopy deck about X") — that intent is already captured.
Otherwise, before dispatching the builder, fire AskUserQuestion with the exact shape below — the question text MUST end with the 7-slug list so the "Other" path is discoverable to the user. Do not paraphrase it; the slug enumeration is the load-bearing part.
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- evals/regression-cases.md 9.6 KB
- recipes/general-html.md 3.9 KB
- recipes/marp-decks.md 4.4 KB
- recipes/pandoc-reports.md 3.5 KB
- scripts/build-marp.sh 2.7 KB runs code
- scripts/build-pandoc.sh 3.2 KB runs code
- scripts/render-mermaid-inline.sh 6.3 KB runs code
- scripts/validate-mermaid.sh 2.1 KB runs code
- templates/base.html 820 B
- templates/deck.md 205 B
- templates/report.css 1.3 KB
- templates/report.html 411 B
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.
- yesterday First seen · 93 lines · 205 tokens per session scan C 5a8274f7f1ea
create is a skill published in the GitHub repository quantumentangled-dev/ai-docs-plugin (6 stars, last pushed 3mo ago), licensed MIT. It adds 205 tokens to every session and 1,521 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…