6.4 KiB
| title | intro | versions | topics | shortTitle | type | |||
|---|---|---|---|---|---|---|---|---|
| About Copilot in GitHub Support | {% data variables.product.prodname_copilot_in_support %} can help you by providing answers to a wide range of questions about {% data variables.product.prodname_dotcom %}'s products and features. |
|
|
About Copilot in GitHub Support | rai |
About {% data variables.product.prodname_copilot_in_support %}
{% data variables.product.prodname_copilot_in_support %} is a new conversational AI tool built to offer immediate assistance within the {% data variables.contact.github_support %} portal, allowing you to tackle issues without waiting for a response from {% data variables.contact.github_support %}. {% data variables.product.prodname_copilot_in_support %} uses a large language model, trained on publicly available {% data variables.product.prodname_dotcom %} documentation, to provide a conversational way to get help for your {% data variables.product.prodname_dotcom %} inquiries without having to wait for customer support to respond to your inquiry.
The primary supported language for {% data variables.product.prodname_copilot_in_support %} is English.
How {% data variables.product.prodname_copilot_in_support %} works
{% data variables.product.prodname_copilot_in_support %}'s functionality can be broken down into a few steps.
Input processing
The input prompt from the user is pre-processed by {% data variables.product.prodname_copilot_in_support %} to determine whether it falls within the feature's remit and, if it does, sent to a large language model to generate a response based on the input. User input should take the form of a plain language description of their query. The system is only intended to respond to {% data variables.product.prodname_dotcom %}-related questions.
Language model analysis
The input prompt is then passed through the {% data variables.product.prodname_copilot_in_support %} language model, which analyzes the input prompt.
Response generation
The language model generates a response based on its analysis of your input. This response will take the form of a plain language answer to your query. Depending on your question, {% data variables.product.prodname_copilot_in_support %} may include steps you can take to try to resolve the issue.
Output formatting
The response generated by {% data variables.product.prodname_copilot_in_support %} is formatted and presented to the user. {% data variables.product.prodname_copilot_in_support %} may use indentation and other formatting features to add clarity to the generated response. Depending upon the type of question from the user, links to relevant content in {% data variables.product.prodname_dotcom %} documentation that the model used to generate its response will also be provided.
Limitations of {% data variables.product.prodname_copilot_in_support %}
Depending on factors such as your input prompt and the complexity of your issue, you may experience different levels of performance when using {% data variables.product.prodname_copilot_in_support %}. The following information is designed to help you understand system limitations and key concepts about performance as they apply to {% data variables.product.prodname_copilot_in_support %}.
Limited scope
{% data variables.product.prodname_copilot_in_support %} is only intended to answer questions related to {% data variables.product.prodname_dotcom %}'s products. {% data variables.product.prodname_copilot_in_support %} is not designed to provide general programming assistance. In addition, it may not be able to help with more complex issues related to your {% data variables.product.prodname_dotcom %} account, for example, resolving billing issues. If {% data variables.product.prodname_copilot_in_support %} cannot help with a GitHub-related issue, you will be prompted to open a ticket with our support team.
Inaccurate responses
The responses you receive from {% data variables.product.prodname_copilot_in_support %} are generated by AI without human intervention and, as such, may not always be accurate. Ensure the answer you have received is accurate by checking the official {% data variables.product.prodname_dotcom %} documentation.
Information only assistance
Currently, {% data variables.product.prodname_copilot_in_support %} cannot take action on your behalf, for example, by changing your settings. {% data variables.product.prodname_copilot_in_support %} can only provide steps for you to configure or troubleshoot a {% data variables.product.prodname_dotcom %} product based on {% data variables.product.prodname_dotcom %} official documentation.
Best practices and use cases
{% data variables.product.prodname_copilot_in_support %} is intended to answer your query related to {% data variables.product.prodname_dotcom %} products from {% data variables.product.prodname_dotcom %} documentation. More information provided in your input increases {% data variables.product.prodname_copilot_in_support %}'s likelihood of providing a relevant and helpful response that may also include best practices, tips, and solutions. {% data variables.product.prodname_copilot_in_support %} consolidates information from multiple {% data variables.product.prodname_dotcom %} documents into a tailored response, and may reduce the time required to answer your question. If you have a follow-up question or another {% data variables.product.prodname_dotcom %}-related query, you can continue using {% data variables.product.prodname_copilot_in_support %}.
{% data variables.product.prodname_copilot_in_support %} also includes links to any documentation it uses when crafting an answer so that you're able to review the material or bookmark it for future reference. As with every large language model, it can make mistakes so please review the information provided to ensure it is free of errors.
Sharing feedback about {% data variables.product.prodname_copilot_in_support %}
We greatly value your feedback. As we continuously improve {% data variables.product.prodname_copilot_in_support %}, we value knowing which answers were helpful and which were not. To let us know, just click the thumbs down under the corresponding response from {% data variables.product.prodname_copilot_short %}.