START HERE FEATURE
Installation
The first useful Semitexa experience should end with a running app and an operator shell you can trust, not with a half-finished checklist.
Feature Guide
A quick orientation block that answers the essential questions: what this feature does, how it works, why it matters, and the key concepts behind it.
What this does
Create the project, review the baseline env contract, and bring up the Semitexa runtime the supported way.
Canonical Install
Create the project and get to a trustworthy first boot
Use the installer, review the shared env baseline, boot the runtime with the Semitexa CLI, and verify the project shape before you start authoring modules.
Step 1
Run the installer
The supported install path is the one-line installer. Use the named-directory form only when you want a specific folder immediately.
curl -fsSL https://semitexa.com/install.sh | bash
curl -fsSL https://semitexa.com/install.sh | bash -s my-project
Step 2
Review the env boundary
Semitexa now writes `.env.default` as the committed baseline. Edit `.env` only when this machine needs overrides such as a different port.
cd my-project
$EDITOR .env
Step 3
Boot and verify
Use the operator shell to start, check, and inspect the runtime instead of guessing from container state alone. After boot, open the local app in the browser and confirm you have a real page, not just a running container.
bin/semitexa server:start
bin/semitexa self-test
bin/semitexa routes:list --json
Default Runtime
- Semitexa uses Docker as the supported local runtime boundary.
- The default app URL is http://localhost:9502 unless `SWOOLE_PORT` is overridden in `.env`.
- If you installed Semitexa Demo during setup, open http://localhost:9502/demo after boot to inspect working feature pages immediately.
- You do not need host PHP or Composer for the normal local flow.
Verified against Semitexa Ultimate 2026.09.19.1020
Installation
Installation in Semitexa should end with a running runtime and a trustworthy operator shell, not with half-finished setup notes.
Canonical flow
- Create the project with the supported installer.
- Move into the project directory.
- Review
.env.defaultas the shared baseline. - Create
.envonly when this machine needs local overrides. - Start the runtime with
bin/semitexa server:start. - Verify the runtime with
self-test, route inspection, and a real browser page load.
Commands
curl -fsSLo install.sh https://semitexa.com/install.sh
# verify checksum/signature from release notes before execution
bash install.sh
bash install.sh my-project
cd my-project
bin/semitexa server:start
bin/semitexa self-test
bin/semitexa routes:list --json
What to verify after boot
- the application responds in the browser
bin/semitexa self-testreports a healthy runtime- route discovery is visible through
routes:list - the project shape is inspectable before you start authoring modules
Why this matters
If the first boot path is ambiguous, every later problem becomes harder to diagnose. Semitexa should feel operationally legible from the first hour.
Boot Verification
What a clean first boot should prove
After `bin/semitexa server:start`, you should be able to verify the runtime from both the browser and the operator shell without guessing what happened.
Open `http://localhost:9502` and confirm the app responds with the starter page instead of a Docker, proxy, or browser error.
If Semitexa Demo was installed, open `http://localhost:9502/demo` and confirm the demo home renders before you explore deeper pages.
Run `bin/semitexa self-test` and expect a clean health check before you start debugging application code.
Run `bin/semitexa routes:list --json` so route discovery is visible instead of assumed.
Use `bin/semitexa ai:ask project --json` when you need to inspect what the scaffold and installed modules actually registered.