MTA-5378 Initial Configurations for Developer Lightspeed #174
Conversation
Pkylas007
changed the title
Pkylas007
changed the title
Pkylas007
changed the title
Pkylas007
changed the title
Pkylas007
changed the title
Pkylas007
changed the title
Pkylas007
force-pushed
the
mta-5637-initial-configuration
branch
from
fb6c9e9 to
c019c9b
Compare
There was a problem hiding this comment.
@Pkylas007 I read over and think the content is looking good.
I do want you to be aware of a potential need to distinguish between "MTA" and "MTA with Developer Lightspeed". Or to be more accurate, I am not positive a change is needed in what you wrote, more I wanted to be sure you understand that we are serving 2 use-cases with this single extension.
We need to consider that the single IDE extension we publish to the Marketplace will satisfy 2 use-cases. We can call the use-cases:
- "MTA" - The basic use-case of static code analysis
- "MTA with Developer Lightspeed" - The generative AI use-case of working with a LLM to get a code suggestion/fix for a specific analysis issue.
I assume we can publish one document to cover both use-cases, we may need to change up a little where we refer to "MTA" vs "MTA with Developer Lightspeed".
The rough idea is consider static code analysis is "MTA", when we are working with a LLM its "MTA with Developer Lightspeed".
| |Solution Server:enabled|Enable the Solution Server client ({ProductShortName} extension) to connect to the Solution Server to perform analysis. | ||
| |Analyze on save|Enable this setting for {mta-dl-plugin} to run an analysis on a file that is saved after code modification. This setting is enabled automatically when you enable Agentic AI mode. | ||
| |Agent mode|Enable the experimental Agentic AI flow for analysis. {mta-dl-plugin} runs an automated analysis of a file to identify issues and suggest resolutions. After you accept the solutions, {mta-dl-plugin} makes the changes in the code and re-analyzes the file. | ||
| |Super agent mode| |
There was a problem hiding this comment.
I am unsure what this mode is, this could be a by-product of an older build.
I do not see this configuration option present in latest extension code:
https://github.com/konveyor/editor-extensions/blob/main/vscode/package.json#L456
| |Excluded diagnostic sources|Add diagnostic sources in the `settings.json` file. The issues generated by such diagnostic sources are excluded from the automated Agentic AI analysis. | ||
| |Get solution max effort|Select the effort level for generating solutions. This can be adjusted depending on the type of incidents. Higher values increase processing time. | ||
| |Get solution max LLM queries|Specify the maximum number of LLM queries made per solution request. | ||
| |Get solution max priority|Specify the maximum priority level of issues to be considered in a solution request. |
There was a problem hiding this comment.
There are 3 settings here which we would like to remove, you can reference the below issue tracking the removal.
konveyor/editor-extensions#697
The 3 settings relate to an older configuration for the agents we called a "solution effort".
This will be for "Get solution max effort", "Get solution max LLM queries", "Get solution max priority". Those 3 settings are no longer used, but they do still exist in code until we address the above issue 697
| ==== | ||
| You must configure either target or source tehcnologies before running an analysis. | ||
| ==== | ||
| |Set rules|Enable default rules and Select your custom rule that you want {mta-dl-plugin} to use for an analysis. You can use the custom rules in addition to the default rules. |
There was a problem hiding this comment.
I am not sure, but worth considering we may want to refer to the Analysis capabilties as "MTA" so that the doc portion for that is covering both "MTA" and "MTA with Developer Lightspeed". Then use the "MTA with Developer Lightspeed" just for the Generative AI portions.
JIRA
Version
Preview