qiankun is a JavaScript framework for building web applications from independently developed micro frontends, allowing multiple teams to combine apps made with different frameworks. It addresses integration, isolation, deployment independence, runtime sandboxing, performance, and shared dependencies between those smaller applications. Catalogue add-ons provide workflows for using qiankun.
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 skills add umijs/qiankun --skill qiankungit clone --depth 1 https://github.com/umijs/qiankunWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/umijs/qiankun/qiankun)<a href="https://agentmods.dev/skills/umijs/qiankun/qiankun"><img src="https://agentmods.dev/badge/skills/umijs/qiankun/qiankun/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/umijs/qiankun/qiankun"><img src="https://agentmods.dev/badge/skills/umijs/qiankun/qiankun.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What 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 5.1 | $0.00069 | $0.00824 |
| Opus 5.5 | $0.00028 | $0.00330 |
| Sonnet 5.5 | $0.00014 | $0.00165 |
| Haiku 4.5 | $0.00007 | $0.00082 |
Grade A, and why
qiankun scanned grade A with 0 findings 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 2d ago.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
qiankun
qiankun is a micro-frontend framework: a main (host) app loads sub (micro) apps from their HTML entries at runtime, each inside a JS sandbox, with opt-in CSS isolation. This skill is the agent-facing manual for qiankun 3.x.
Task routing
Read only the reference file(s) the task at hand needs:
| Task | Read |
|---|---|
| Create a micro (sub) app, or convert an existing Vite app | references/create-micro-app.md |
| Create a main (host) app | references/create-main-app.md |
| Anything else — migration, style isolation, sandbox, debugging | https://qiankun.umijs.org and examples/ in https://github.com/umijs/qiankun |
Before you start
Determine these, asking the user only when not inferable from context:
- App type — main (host/shell) or sub (micro app).
- App name — lowercase letters, numbers, hyphens. The name the host registers must exactly match the name the sub app uses in its classic-mode fallback.
- Framework — React or Vue; TypeScript or JavaScript. Templates in the references are TS; strip types for JS.
- Dev port — must be fixed and unique per app, because the host hard-references it in
entry. Convention: main app7099, sub apps7101,7102, …
Package versions — until qiankun 3.0 stable ships, rc is the dist-tag for the core packages:
| Package | Dist-tag | Goes in |
|---|---|---|
qiankun |
rc |
main app dependencies |
@qiankunjs/react / @qiankunjs/vue |
rc |
main app dependencies (optional UI binding) |
@qiankunjs/bundler-plugin |
rc |
sub app devDependencies |
Key facts
These invariants shape every task below:
- Vite sub apps need no dedicated build mode. qiankun 3 loads
<script type="module">natively through its ESM sandbox — regularvite dev/vite buildoutput is qiankun-ready as-is. - The JS sandbox is on by default. CSS isolation is opt-in per app via
styleIsolation: true(runtime CSS@scope). - A sub app must stay runnable standalone: when not loaded by qiankun (
window.__POWERED_BY_QIANKUN__is undefined) it renders itself directly. - Always unmount what you mount.
loadMicroAppreturns a handle whose.unmount()must be called when the app leaves; theMicroAppcomponent bindings do this automatically on component unmount. Unmounted instances stay cached for reuse by app name; only.unload()/unloadMicroApp(name)releases them, and the bindings never call those. - Webpack sub apps use
@qiankunjs/bundler-plugin/webpackinstead of the Vite plugin; see the docs.
What ships with it
2 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.
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.
- 2d ago Changed 68bc884c95d0
- 1mo ago First seen · 51 lines · 69 tokens per session scan A 820936d95dee
qiankun is a skill published in the GitHub repository umijs/qiankun (16,697 stars, last pushed 2d ago), licensed MIT. It adds 69 tokens to every session and 824 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
vercel-microfrontends
Guide for building, configuring, and deploying microfrontends on Vercel. Use this skill when the user mentions microfrontends, multi-zones, splitting an app across teams, independent deployments, cross-app routing, incremental migration, composing multiple frontends under one domain, microfrontends.json…
create-pr
Creates a GitHub PR with a Linear-ticket-prefixed title and a decision-led, narrative description for Prisma 8. Use when the user wants to create a pull request, open a PR, or submit changes for review.
review-implement-phase
Implements triaged review actions, commits focused fixes, and posts Done plus resolves threads. Use when the user wants only the implementation phase of the review-framework workflow.
no-bare-casts
Writing as in TypeScript or TSX production code, modifying a file that contains a bare as cast, silencing a type error with a cast, encountering as unknown as, or reviewing a cast site.
write-architecture-docs
Write or rewrite architecture subsystem docs, ADRs, and reference material for the engineering team. Use when creating, updating, or reviewing docs under docs/architecture docs/, or when the user asks you to write documentation that describes the system's design.
compiler-port
Port a compiler pass from TypeScript to Rust. Gathers context, plans the port, implements in a subagent with test-fix loop, then reviews.