From b1be24533fe9e7cc640e06382e28ae6568b69b92 Mon Sep 17 00:00:00 2001 From: Alex Ross <38270282+alexr00@users.noreply.github.com> Date: Thu, 18 Dec 2025 07:35:44 +0100 Subject: [PATCH] Add upvotes to issue-grouping prompt (#283850) --- .github/prompts/issue-grouping.prompt.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/.github/prompts/issue-grouping.prompt.md b/.github/prompts/issue-grouping.prompt.md index e4fa31d0327..8f6bfea7660 100644 --- a/.github/prompts/issue-grouping.prompt.md +++ b/.github/prompts/issue-grouping.prompt.md @@ -16,6 +16,7 @@ tools: a. Using the GitHub MCP server, fetch only one page (50 per page) of the open issues for the given assignee and label in the `vscode` repository. b. After fetching a single page, look through the issues and see if there are are any good grouping categories.Output the categories as headers to a local file categorized-issues.md. Do NOT fetch more issue pages yet, make sure to write the categories to the file first. 2. Repeat step 1 (sequentially, don't parallelize) until all pages are fetched and categories are written to the file. -3. Use a subagent to Re-fetch only one page of the issues for the given assignee and label in the `vscode` repository. Write each issue into the categorized-issues.md file under the appropriate category header with a link. If an issue doesn't fit into any category, put it under an "Other" category. +3. Use a subagent to Re-fetch only one page of the issues for the given assignee and label in the `vscode` repository. Write each issue into the categorized-issues.md file under the appropriate category header with a link and the number of upvotes. If an issue doesn't fit into any category, put it under an "Other" category. 4. Repeat step 3 (sequentially, don't parallelize) until all pages are fetched and all issues are written to the file. -5. Show the categorized-issues.md file as the final output. +5. Within each category, sort the issues by number of upvotes in descending order. +6. Show the categorized-issues.md file as the final output.