https://github.com/lila-team/lila
Official CLI for Lila: the testing framework for startups
https://github.com/lila-team/lila
ai-agents llm qa qa-automation startups testing
Last synced: 11 months ago
JSON representation
Official CLI for Lila: the testing framework for startups
- Host: GitHub
- URL: https://github.com/lila-team/lila
- Owner: lila-team
- Created: 2025-01-16T00:50:37.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-05-08T19:33:06.000Z (about 1 year ago)
- Last Synced: 2025-05-08T19:44:09.944Z (about 1 year ago)
- Topics: ai-agents, llm, qa, qa-automation, startups, testing
- Language: Python
- Homepage: https://lila.dev
- Size: 197 KB
- Stars: 15
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Citation: CITATION.cff
Awesome Lists containing this project
README
[
](https://lila.dev/)
The best testing framework for startups 🚀
[](https://pypi.org/project/lilacli/)
[](https://docs.lila.dev)
[](https://github.com/lila-team/lila/actions/workflows/daily-run.yml)
[](https://twitter.com/lila__dev)
[](https://discord.gg/kZ7TEmxH)
Lila CLI is a powerful tool for running end-to-end tests written in human-readable plain text using YAML files. It simplifies the testing process by allowing anyone in the team to write tests in a natural, easy-to-understand format.
**No coding required.**
## Quick start
With pip (Python>=3.11)
```
pip install lilacli
```
Install Playwright Chromium web drivers
```
playwright install chromium
```
Create your first testcase in a `demo.yaml` file
```yaml
steps:
- goto: https://www.google.com/maps
- input: Empire State on the main search bar
verify: a dropdown with suggestions appears
- click: on the first suggestion from the dropdown
verify: the map shows the Empire State Building in NY
```
Fetch an API Key from [Lila app](https://app.lila.dev) and add it to an `.env` file:
```
LILA_API_KEY=...
```
Run
```
lila run demo.yaml
```
## How does it work?
Lila runs your app in a **local browser** with [Playwright](https://playwright.dev/python/) and uses an LLM-powered engine to guide the CLI run high level instructions. Lila extracts information from the DOM using the awesome library [browser-use](https://github.com/browser-use/browser-use)
### Why Lila?
* No coding required.
* Self healing tests, does not rely on low level implementation.
* Anyone in the team can implement tests.
* Integrates natively with Playwright storage states.
* Runs browser locally, making it ideal for localdev or staging environments.
### Writing tests
Tests are just YAML files. Here is the same example as the quick start.
```yaml
steps:
- goto: https://www.google.com/maps
- input: Empire State on the main search bar
verify: a dropdown with suggestions appears
- click: on the first suggestion from the dropdown
verify: the map shows the Empire State Building in NY
```
### Step types
**goto**: navigate to a URL
```yaml
- goto: https://private.staging.my-app.com
```
**click**: perform a click on a described element
```yaml
- click: on the checkout cart
# or
- click: on the direction item
# or
- click: on button 'Create'
```
**input**: type text into an input field
```yaml
- input: foo@bar.com as email
# or
- input: iPad mini in the main search bar
```
**submit**: submits a form by clicking a button or pressing enter
```yaml
- submit: the login form
# or
- submit: the search
```
**wait**: just waits N seconds
```yaml
- wait: 10 # waits 10 seconds
```
**pick**: pick an element from a dropdown
```yaml
- pick: EST from the timezone dropdown
```
**exec**: run a bash command
```yaml
- exec: | # multiline
curl -X POST ...
# or
- exec: psql -c "INSERT ..." # single line
```
### Verifications
Each step can have one or more `verify` assertions.
```yaml
- goto: https://my-app.com/login
verify: there is a login form with username and password # Single verification
- click: the checkout cart
verify: # multi verifications
- 5 items appear in the purchase summary
- the total is 500 USD
```
### More on building tests
For more information on how to build tests, [checkout the guides](https://docs.lila.dev/guides/intro)
### More examples
For test examples, Lila runs daily a suite of tests over public URLs. Check them out [here](https://github.com/lila-team/examples)
## Creating an account
Lila is free for beta users. Create your free account at [https://lila.dev](https://lila.dev).
You will need the account to fetch an API key to run lila CLI.
## Reusing state
Lila runs save the browser state in the output folder. To re-use where another test left off, just pass `--browser-state` parameter.
For example:
**login.yaml**:
```yaml
steps:
- goto: https://my-app.com/login
- input: foobar as username
- input: barbaz as password
- submit: the login form
verify: login was successful
```
`lila run login.yaml` -> This will generate `lila-output/login.json`
**dashboard.yaml** (needs a logged in user)
```yaml
steps:
- goto: https://my-app.com/dashboard
verify: there is a panel displaying metrics
```
`lila run dashboard.yaml --browser-state lila-output/login.json`
## CLI Parameters
* `--browser-state`: Initialize the browser with a Playwright JSON storage session
* `--tags`: Filter tests by specified comma-separated tags
* `--exclude-tags`: Filter out tests by specified comma-separated tags
* `--output-dir`: Specify the output directory for the browser states
* `--config`: Point to the `lila.toml` config file if not using the root location
* `--headless`: If present, does not open browser
* `--debug`: Displays debug logs
## CI/CD ready
Checkout [an example for a daily run using GHA](.github/workflows/daily-run.yaml)
## Documentation
For comprehensive documentation, visit [docs.lila.dev](https://docs.lila.dev).
## Examples
Find a variety of test examples in our [examples repository](https://github.com/lila-team/examples), showcasing different testing scenarios and best practices.
## Contact
Reach out at info@lila.dev
Join our [Discord server](https://discord.gg/6rRfZUqh)
Follow us at [Twitter - X](https://x.com/lila__dev)