maestro-reporter 0.8.0

py-maestro-reporter

py-maestro-reporter is a lightweight tool that helps you:

• Run Maestro test seamlessly
• Parse Maestro JUnit reports
• Send summarized test results to a Lark/Slack respective group

It can be used either as a CLI tool or a python package into your own test pipelines

Prerequisites

• Python 3.10 or above
• Maestro framework installed on your system (version 2.0.0 or above)
• Device/emulator with the app under test installed
• Lark/Slack webhook URL (optional)

Installation

For the installation, you can either install the package from PyPI or from source. If you'd like to install from PyPI, you can install it with:

pip install maestro-reporter

Or, using uv:

uv pip install maestro-reporter

Or, if you'd prefer to install from source, you'll need to clone this repository and install it in editable mode:

pip install -e .

Usage

Using as a CLI argument

This package expose a CLI via the reporter module

All you need to do is, ensure you have Maestro installed on your device, example flows which is going to be tested, physical device / emulator and Webhook URL from Lark. Once you have all of these, you can run with :

python -m reporter \
    -c "maestro test examples/facebook-sign-up-flow.yaml --format junit --output tests/report.xml" \
    -r "tests/report.xml" \
    -w "https://webhook.url.com"

Parsing an existing report

Or, if you only want to run and parse the report without testing, you can use --no-run flag

python -m reporter \
    --no-run \
    -r "tests/report.xml" \
    -w "https://webhook.url.com"

You can also overrides the Webhook URL by setting the LARK_URL or SLACK_URL environment variable in your .env file, depending on the provider you choose.

Using the reporter package

Otherwise, if you'd like to run the tests without using the CLI arguments and you need to run the tests with the reporter package, you can follow the example below:

# this will test the Facebook sign-up flow
import os
from dotenv import load_dotenv
from reporter import parse_xml_report, send_report_to_lark, run_maestro_command


load_dotenv()


command = "maestro test examples/facebook-sign-up-flow.yaml --format junit --output build/maestro-results/report.xml"
report_path = run_maestro_command(command=command, cwd="tests")
parsed_result = parse_xml_report(file_path=str(report_path))
report = send_report_to_lark(
    summary=parsed_result,
    title="Maestro Reporter Test",
    color_template="Green",
    webhook_url=os.getenv("LARK_URL"),
)

The parameters of color_template and title are mandatory, if you don't provide them, the default values will be used

Note (v0.8.0): run_maestro_command reads the --output argument from your command, creates the report's parent directory automatically, and returns the resolved report path. When Maestro exits with a non-zero code, the report is still returned if it exists, so failed-test results can be reported. RuntimeError is raised only when no report file is found. run_maestro_command and parse_xml_report raise an exception on failure instead of silently returning None.

All successful tests will be displayed in the log stream handler, for example:

27-11-2025 : 10:51:46 : main : [WARNING] : No color template provided, using default color template or you can set it with `--color` flag
27-11-2025 : 10:51:46 : main : [WARNING] : No title provided, using default title or you can set it with `--title` flag
27-11-2025 : 10:51:46 : main : [INFO] : --no-run flag is set, skipping Maestro tests
27-11-2025 : 10:51:46 : main : [INFO] : Parsing Maestro report file: tests/report.xml
27-11-2025 : 10:51:46 : main : [INFO] : Sending Maestro report to Lark...
27-11-2025 : 10:51:46 : reporter.sender : [INFO] : Lark message sent successfully
27-11-2025 : 10:51:46 : main : [INFO] : Maestro report sent successfully

Once the report is sent successfully, you should be able to see the interactive card message in your Lark group like the following image

Otherwise, if you want to use Slack as a reporting platform, the card message will be displayed as follows

CLI arguments

List of available CLI arguments that you can use with this package:

Notes

• At the moment, this package only supports the parsing of the junit format as follows for the Maestro report
• In addition, the webhook integration currently supports Lark and Slack
• The interactive card message is built using the msg_actioncard message type for Lark and Block Kit for Slack

Further references

• Generate report with Maestro
• Setup Lark Webhook URL in Lark group