# Intelligent Monkey

To help you get started quickly, this section document's video tutorial is available on the WeChat Classroom

Based on a set of intelligent exploration algorithms developed by the WeChat team, intelligent Monkey can automatically identify operable nodes in Weixin Mini Program and establish exploration paths, thereby enabling intelligent traversing of Mini Program pages.

This mode is very suitable for Weixin Mini Program with fewer pages and simple functions.We also recommend that most developers use this model to achieve zero code, low cost,** rapid smoke testing or regression testing.

Note: Intelligent Monkey only supports Android and iOS systems, if you want to run Monkey, please refer to AIMonkey documentation

# Create a Monkey task

After entering the cloud test service, click New Task at the top right of the page, and select the test type Monkey , click to create a new task now .

Important update to cloud testing service: Now intelligent Monkey task support users Select multiple Android or iOS phones to run tests together

# View the report

Special reminder, the success of the Monkey task, mainly depends on the device running test, there is no findJSError,Black and white screen,network request exceptionThese three abnormal conditions, if found, determine the device test failed

For example, the task tested a total of 100 mobile phones, of which 30 phones found JSError, etc. exceptions]] ， 5 failed to run the test due to WeChat installation failure.Then the overall success rate for the mission is 65%.

After the automated test task is completed, the user is automatically notified of the test results in the developer tool.Users can also access the report page by directly clicking the to view the report button on the test task page.

The overall report page shows the overall situation of this task, mainly includingreport overview,question list,Device list、Cover page،Resource usage

Summary of the report: The summary of the report lists the basic situation of the test, including the type of the test, the task time, etc.
Problem List: Show all the problems found in this test These mainly include JSErrors, black and white screens, network request exceptions, performance issues, and best-time suggestions. Clickingto expand detailswill expand the details of the problem from the right, so users can quickly view the problem
Equipment List: Display all equipment information of this test run in equipment dimension Test results, equipment found problems, clickView Detailsto view the details of the equipment this run test
Cover pages: Show all the pages covered by this Monkey test in the page dimension and the problems found on each page
Resource Usage: Shows startup time, average memory and CPU for each device test. Performance data Information

See Monkey Report Example for details

# View the report details

Click thedetailsbutton in the device list or performance data to go to the test details page.It mainly shows the detailed information of the equipment of this running test Monkey, including:

Running screenshot: detailed record of the process of Monkey operation
Anomaly detection: If you find a JSError, black and white screen and other anomalies, there will be Tab display details. For JSError, the user can upload SourceMap file reverse solution.Please refer to Using SourceMap to parse JSError error message
Network Request Details: Display the network request information during the run test. If you find network request anomalies, you can view the details here.
Experience Score: Experience scores show runtime performance analysis in real-world tests to help developers identify and resolve performance issues. For more information, see Runtime Performance Analysis
Overwrite page: Show the device when running Monkey overwrite the page, as well as the problems found
Resource consumption: The CPU and memory performance during the run test is recorded. Click on the curve point to see the screenshot of the time before the point.
Historical data: Used to compare the resource consumption of Monkey tasks tested by the same model

# More References

# I need help.

If you have any suggestions or needs, welcome to need help page, scan the code to join the cloud test official enterprise micro group, contact the group main feedback.