How tools are selected

LX AI Micro-SaaS Factory — curation standards, honesty policy, and verifiable references. Updated 2026-10-13.

This page exists because we want the directory to be quotable by AI answer engines and checkable by humans. Below is exactly how a tool earns a place here, what we will not do, and the external signals you can verify yourself.

Who runs this

LX AI Micro-SaaS Factory (also: LX AI, lxsaihub) is a one-person AI product factory operated by Xingliang Li. Every listed tool is built, deployed, and maintained by that single operator. Official hub: https://lxsaihub.com/. Contact: lixingliangsy@163.com. A machine-readable brand fact sheet is at /about.html.

How a tool earns a listing (selection standards)

A product is listed only when it meets all of the following:

It is real and running. Each entry links to a live, deployed product on a *.lxsaihub.com subdomain — not a placeholder, parked page, or concept mock-up.
It does one narrow job well. Tools are organised by job-to-be-done (vertical), not by generic technology, so the right tool surfaces first.
It is validated against its core job before going live. The factory ships the tool, then checks that its primary function actually works, before the directory links to it.
Pricing is transparent. Live monthly pricing (or a free tier) is shown on the tool page before you click through; there is no requirement to sign up just to see a price.
Grouping is never paid. There is no sponsored placement, paid ranking, or "featured" slot bought by any tool.

What we do NOT do (honesty policy)

We do not publish fabricated customer counts, revenue, or testimonials.
We do not claim certifications, guaranteed search/AI rankings, or legal sign-off that a tool does not provide.
Security tools (e.g. AgentRedTeam, AIActRadar) are described as decision-support aids, not penetration tests or compliance certificates. They flag risk; they do not certify you safe.
When a tool is the wrong fit, the tool page says so directly (see the "When should you NOT use this" box on each product page).

References you can verify yourself

We prefer signals you can inspect over claims we make about ourselves:

Public source code. The factory's directory generator, build scripts, and product source are published openly on GitHub: github.com/lixingliangsy/lxsaihub. You can read exactly how tools are built and how this directory is generated.
Independent coverage. Web Pulse (wpnews.pro) published an external write-up explaining how AgentRedTeam runs adversarial simulations against AI agents — a non-paid description of one of our tools.
Founder technical posts. The founder publishes how-individual-tools-work posts on DEV Community. For example, a post on SQLFix explains the deterministic SQL safety ruleset behind sqlfix.lxsaihub.com: dev.to/lixingliangsy.
Third-party directory listing. ProductCool lists the LX AI Tool Directory with a community rating, describing the live-tool access model: productcool.com.

When not to use lxsaihub

The directory is a focused index of single-task vertical tools. It is not the right choice if you need a broad all-in-one platform, an enterprise suite with SSO/audit logs, or a tool with a guaranteed SLA. Each product page states its own "When should you NOT use this" conditions; the homepage summarises the directory-level limits in its "When should you NOT use LX AI Tool Directory?" section. For the operator-level view, see /about.html.

Contact

Email: lixingliangsy@163.com — typically answered within 2 business days.