feedback-to-issues_skill
- TypeScript
19.7k
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill dyad-sh/dyad --skill feedback-to-issues- SKILL.md3.7 KB
Overview
This skill turns customer feedback (email, ticket, or text file) into discrete GitHub issues, detects duplicates, proposes new issues for review, creates approved issues, and drafts a customer reply. It streamlines triage by extracting actionable items, searching the repository for matching issues, and preparing clear issue bodies and labels for maintainers.
How this skill works
The skill parses the feedback into separate, actionable items with an imperative title, type (bug/feature/improvement/question), a short description, priority, and relevant quotes. It searches GitHub for potential duplicates using multiple keyword variations, then presents a three-part report: already filed matches, proposed new issues with bodies and labels, and a summary. After user approval and any edits, it creates the issues and drafts a concise reply email linking to each issue.
When to use it
- Processing customer support emails or helpdesk tickets into actionable engineering work.
- Triage after product launch to capture early user reports and feature requests.
- Regularly converting qualitative feedback from sales or success teams into tracked issues.
- When you need a fast, consistent way to check for duplicate issues before filing new ones.
Best practices
- Feed the raw message body or a plain text file to preserve original quotes for context.
- Review and edit proposed titles, priorities, and bodies before approval to match team conventions.
- Use multiple, focused keyword searches during duplicate detection to reduce false negatives.
- Label new issues with product/component names in addition to type and priority for easier routing.
- Keep reply emails short and link to issues so customers can follow progress themselves.
Example use cases
- A support agent forwards a bug report email to produce a high-priority GitHub issue and reply to the customer.
- Product managers batch-process feature requests gathered from interviews into proposed issues for roadmap review.
- Customer success compiles UX complaints into discrete improvement issues and removes duplicates first.
- Open-source maintainers convert community questions and reproducible bugs into tracked issues with links for reporters.
FAQ
Not until you explicitly approve the proposed issues. It first presents matches and proposals for review, then creates only approved items.
How does duplicate detection work?
It runs multiple GitHub issue searches using varied keywords from each extracted item and reports matching open or closed issues with reasoning so you can confirm duplicates.
