Skip to content

test(#27): test research agent for a feature ticket#82

Draft
andrablaj wants to merge 4 commits into
mainfrom
27-research-testing
Draft

test(#27): test research agent for a feature ticket#82
andrablaj wants to merge 4 commits into
mainfrom
27-research-testing

Conversation

@andrablaj

Copy link
Copy Markdown
Member

Description

Closes #27

Code review checklist

  • Readable: Concise, well named, follows the style guide, documented if necessary.
  • Documented: Configuration and user documentation on cht-docs
  • Tested: Unit and/or e2e where appropriate
  • Backwards compatible: Works with existing data and configuration or includes a migration. Any breaking changes documented in the release notes.

License

The software is provided under AGPL-3.0. Contributions to this project are accepted under the same license.

@andrablaj

Copy link
Copy Markdown
Member Author

1. Testing a feature ticket

Screenshot 2026-03-19 at 17 33 20

@andrablaj

Copy link
Copy Markdown
Member Author

@sugat009, before I move forward with other tickets, can you please have a look at the files I created and let me know if this is what you had in mind as an output for each ticket testing? I will need help from the team to validate the accuracy of the research results, but I can do that at a later stage, once I get the format right. Thanks!

@sugat009

Copy link
Copy Markdown
Member

@andrablaj, yes, that is the expected output!

I noticed in line 45 that the Kapa AI MCP isn’t currently configured. To get higher-quality research results, you can enable it by adding this to your .env file:

MCP_SERVER_URL=http://localhost:3000/mcp

Note: I’ve also opened a ticket to improve the research quality further by breaking prompts into smaller chunks before they hit the Kapa AI server. If the current output feels slightly off, don't worry, more optimizations are on the way!

@Hareet

Hareet commented Mar 24, 2026

Copy link
Copy Markdown
Member

@andrablaj, yes, that is the expected output!

I noticed in line 45 that the Kapa AI MCP isn’t currently configured. To get higher-quality research results, you can enable it by adding this to your .env file

I think this should be configured out of the box. Thoughts?

@sugat009

Copy link
Copy Markdown
Member

@andrablaj, yes, that is the expected output!
I noticed in line 45 that the Kapa AI MCP isn’t currently configured. To get higher-quality research results, you can enable it by adding this to your .env file

I think this should be configured out of the box. Thoughts?

Out of the box as in having a default value point towards the dev Kapa AI: https://mcp-docs.dev.medicmobile.org/mcp?

@Hareet

Hareet commented Mar 24, 2026

Copy link
Copy Markdown
Member

Out of the box as in having a default value point towards the dev Kapa AI: https://mcp-docs.dev.medicmobile.org/mcp?

Yup, that should result in the following types of scenarios:

  • User with regular coding agent, configured with nothing, goes and tries to answer the user's research query
  • User running a cht-agent research query, doesn't have to worry about anything, and it's already pre-configured to give a better answer than regular coding agent with no docs mcp.

@sugat009

Copy link
Copy Markdown
Member

cool, let me create a ticket for this change.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

Manual test: Research workflow with 5 sample tickets

3 participants