Checking for potential issues in source code by augmenting a natural language prompt with issues that are similar to source code metadata

A computer-program product comprising one or more non-transitory machine-readable storage media, including stored instructions configured to cause a computing system to perform a set of actions including:

• generating a particular vector embedding representing a set of content describing functionality of a set of code;
• generating a plurality of issue vector embeddings representing a plurality of issues based at least in part on sets of natural language describing information about the plurality of issues;
• determining a plurality of distances between the plurality of issue vector embeddings and the particular vector embedding;
• based at least in part on the plurality of distances, selecting, to support communication with a large language model, a subset of the plurality of issues;
• generating a prompt for execution at least in part by the large language model, wherein the prompt comprises:
• at least some natural language describing each issue of the subset of the plurality of issues, and
• at least some content of the set of content;
• prompting the large language model with the prompt;
• accessing a result based at least in part on the execution;
• storing a new association between a particular issue and the set of code based at least in part on the result;
• and
• based at least in part on the new association, causing display of information identifying the particular issue as a candidate issue for the set of code.

A system comprising:

• one or more processors;
• one or more non-transitory computer-readable media storing instructions, which, when executed by the system, cause the system to perform a set of actions including:
• generating a particular vector embedding representing a set of content describing functionality of a set of code;
• generating a plurality of issue vector embeddings representing a plurality of issues based at least in part on sets of natural language describing information about the plurality of issues;
• determining a plurality of distances between the plurality of issue vector embeddings and the particular vector embedding;
• based at least in part on the plurality of distances, selecting, to support communication with a large language model, a subset of the plurality of issues;
• generating a prompt for execution at least in part by the large language model, wherein the prompt comprises:
• at least some natural language describing each issue of the subset of the plurality of issues, and
• at least some content of the set of content;
• prompting the large language model with the prompt;
• accessing a result based at least in part on the execution;
• storing a new association between a particular issue and the set of code based at least in part on the result;
• and
• based at least in part on the new association, causing display of information identifying the particular issue as a candidate issue for the set of code.