moonshot icon indicating copy to clipboard operation
moonshot copied to clipboard

Moonshot - A simple and modular tool to evaluate and red-team any LLM application.

Results 35 moonshot issues
Sort by recently updated
recently updated
newest added

When we run benchmarks on the same endpoint repeatedly in the same runner, the results will be cached. All the tests will receive the same score since they are cached....

enhancement

When the response is in Chinese, CLI version does not print out the entire response. See screenshot below: ![image](https://github.com/aiverify-foundation/moonshot/assets/92433521/93d366d6-4a35-4ee3-834c-c19b5d2404a0) ### Produce Issue 1. Create a new red team session 2....

enhancement

## Description As categories and tags has been added to the property of cookbook, there requires a change in the schema of the cookbook. This PR changes the schema of...

## Description As a dev, I want to test my RAG application on its capability so that I can evaluate how well it performs. Acceptance Criteria Implementation of the following...

## Description Bug is causing view_bookmark not to show an existing bookmark ## Motivation and Context Bug fix ## Type of Change A bug fix ## How to Test 1....

## Description There were remnants of attack module codes in view_cookbook that caused an error while viewing cookbooks. ## Motivation and Context Bug fix ## Type of Change A bug...

## Description Currently, when EKS instance crashes, all endpoints gets deleted and/or resetting to default settings and benchmark runs and sessions gets deleted. ## Motivation and Context Currently, when EKS...

## Description Updated FAQ with suggested solution for utf-8 charmap problem for Windows ## Motivation and Context Update FAQ ## Type of Change Documentation updates ## How to Test ##...

## Description Add unit tests to connectors-endpoints and connectors. This PR will cover changes in MS-632. ## Motivation and Context Add unit tests 1. test_connector_endpoint_arguments.py 2. test_connector_endpoint.py 3. test_connector_prompt_arguments.py 4....

## Description Update retrying mechanism and making user experience better ## Motivation and Context While doing the unit tests for the connector endpoints, there are some key, values that are...