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/abersheeran/kui/kui-frameworknpx skills add abersheeran/kui --skill kui-frameworkgit clone --depth 1 https://github.com/abersheeran/kuiWrote 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/abersheeran/kui/kui-framework)<a href="https://agentmods.dev/skills/abersheeran/kui/kui-framework"><img src="https://agentmods.dev/badge/skills/abersheeran/kui/kui-framework.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00044 | $0.04235 |
| Opus 5 | $0.00022 | $0.02117 |
| Sonnet 5 | $0.00009 | $0.00847 |
| Haiku 4.5 | $0.00004 | $0.00424 |
Grade A, and why
kui-framework 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 5d 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 — 659 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Kui Web Framework Guide
Kui is a Python web framework supporting both ASGI (async) and WSGI (sync). It provides type-safe parameter binding via Annotated, dependency injection, OpenAPI documentation generation, class-based views, and middleware composition.
ASGI imports: from kui.asgi import ...
WSGI imports: from kui.wsgi import ...
The two modules expose nearly identical APIs; ASGI handlers are async def, WSGI handlers are plain def. All examples below use ASGI unless noted.
1. Application
from kui.asgi import Kui, OpenAPI
app = Kui(
routes=[], # Initial routes
http_middlewares=[], # App-level HTTP middleware
socket_middlewares=[], # App-level WebSocket middleware (ASGI only)
cors_config=None, # Optional CORSConfig dict
exception_handlers={}, # {status_or_exc_type: handler}
on_startup=[], on_shutdown=[], # Lifespan callbacks (ASGI only)
templates=None, # Optional Jinja2Templates instance
response_converters={}, # {type: converter_func}
json_encoder={}, # {type: serializer_func}
)
app is a standard ASGI callable (async def __call__(scope, receive, send)).
For WSGI it is a standard def __call__(environ, start_response).
Shared state
app.state.db = create_pool() # Set once; ImmutableAttribute prevents reassignment
2. Routing
2.1 Decorator style (most common)
@app.router.http.get("/users/{user_id:int}")
async def get_user(user_id: Annotated[int, Path()]):
return {"id": user_id}
@app.router.http.post("/users")
async def create_user(body: Annotated[UserCreate, Body()]):
return body, 201
# Any HTTP method
@app.router.http("/health")
async def health():
return "ok"
Available method shortcuts: .get(), .post(), .put(), .patch(), .delete().
Decorator parameters: path, name=, middlewares=[], summary=, description=, tags=.
2.2 Route objects with << operator
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.
- 5d ago First seen · 659 lines · 44 tokens per session scan A e693ee289f22
kui-framework is a skill published in the GitHub repository abersheeran/kui (297 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 44 tokens to every session and 4,235 once invoked, about $0.0002 per session on Opus 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
FastAPI Customer Support Tech Enablement
Comprehensive FastAPI skill for building modern Python web APIs with focus on customer support systems, ticket management, real-time chat, and backend operations.
flask
Flask - Lightweight Python web framework for microservices, REST APIs, and flexible web applications with extensive extension ecosystem.
flask
Flask - Lightweight Python web framework for microservices, REST APIs, and flexible web applications with extensive extension ecosystem.
flask-expert
Expert-level Flask web development, REST APIs, extensions, and production deployment.
cocos-network
实现网络通信时使用,包括 HTTP 请求封装、WebSocket 长连接、协议定义和回调处理。.
FastAPI Modern Web Development
Production-grade FastAPI development with async patterns, Pydantic v2, dependency injection, ML/AI endpoint design, and modern Python best practices for building high-performance REST APIs.