Collect content ratings and issue categories into a local SQLite database.
npx skills add https://github.com/boringdata/kurt-demo --skill feedback-boringdata
Please help me install this Agent Skill. Skill: feedback Source: https://github.com/boringdata/kurt-demo/tree/main/.claude/skills/feedback-skill Command: npx skills add https://github.com/boringdata/kurt-demo --skill feedback-boringdata
Collects user ratings on content quality and categories issues, stores events in a local SQLite database, and surfaces patterns to guide manual rule updates. It provides a lightweight, low-friction way to gather feedback without automating changes.
High-intent search queries and answers about installing and using this skill.
Feedback collection identifies patterns in content issues by rating items 1-5, categorizing problems, and storing them in a local SQLite database. Use the rate subskill to gather feedback, then patterns to surface recurring issues and guide manual rule updates.
Yes. Feedback stores all ratings, issue categories, and comments in a local SQLite database (.kurt/kurt.sqlite), enabling lightweight storage and pattern detection across content items like outlines and drafts without external dependencies.
The dashboard subskill displays feedback trends, distributions, and recent ratings to help you spot quality patterns. Access it via feedback-skill dashboard to review aggregate data without manual log parsing.
Run feedback-skill patterns to surface recurring issues, then apply insights manually using complementary skills like writing-rules-skill. This low-friction workflow lets you validate patterns before automating changes.
No. Feedback creates and manages its own SQLite database automatically. Just run feedback-skill rate <path> to start collecting ratings—no prerequisite configuration or external database tools required.
Yes. Feedback accepts issue category input alongside ratings and optional comments for each content item, enabling you to distinguish problem types—such as clarity, tone, or structure—and surface category-specific patterns.
anderskevStore and query vector embeddings in SQLite using vec0 virtual tables.
Snowflake-LabsA concise summary of your roleplay character: name,</name> personality, and/or behavior, formatted as a single string sentence. Typically不超过50字，此处省略
kevinnftFetch trending GitHub repositories and score their usefulness for adoption.
prassoaiAutomate local code reviews with isolated Git worktrees and streaming issues.
