Identify duplicate GitHub issues by semantically matching titles and bodies.
npx skills add https://github.com/japurcell/skills --skill dedupe
Please help me install this Agent Skill. Skill: dedupe Source: https://github.com/japurcell/skills/tree/main/skills/dedupe Command: npx skills add https://github.com/japurcell/skills --skill dedupe
Find duplicate GitHub issues by semantically comparing a new issue against existing ones.
Input a new issue reference and get back a list of potential duplicates.
High-intent search queries and answers about installing and using this skill.
Semantic duplicate detection compares a new issue against existing ones using titles and bodies. It works across both open and closed issues in a targeted GitHub repository, performing multi-step comparisons with configurable term extraction to surface likely duplicates for triage.
Yes, semantic comparison evaluates both open and closed GitHub issues. By analyzing titles and bodies across the entire repository history, it identifies existing discussions related to a new issue, helping maintainers avoid fragmentation and consolidate related topics.
Using semantic similarity scoring is the best way to triage new GitHub issues for existing discussions. The Skill extracts configurable terms and evaluates similarity across issue titles and bodies to identify potential duplicates, consolidating related conversations effectively.
Semantic issue matching requires accessing issue data such as titles, bodies, and labels. The Skill utilizes this metadata to perform multi-step comparisons with configurable term extraction, evaluating similarity to accurately identify potential duplicates within targeted GitHub repositories.
JuniYadiParse GitHub issue titles and bodies to extract errors, file paths, and function names.
bradtaylorsfSearch GitHub issues and assess similarity to detect duplicates.
NourcineAbAutomate GitHub issue tracking, board synchronization, and sprint planning workflows.
SufficientDaikonDetect and deduplicate text content using normalization, SHA-256 fingerprinting, and Jaccard similarity.
