Hi everyone,

While exploring the BLT platform and its bug reporting workflow, I started thinking about how duplicate submissions might become more common as the number of reports grows.

I was wondering if it would make sense to introduce some form of automated duplicate detection that compares new bug reports with existing ones before submission or during triage.

Some possible approaches that came to mind include:

• Text similarity or fuzzy matching between report titles/descriptions
• Embedding-based similarity (e.g., cosine similarity on report text)
• Keyword extraction and overlap detection

Potential benefits could include:

• Reducing duplicate bug reports
• Lowering the manual triage workload for maintainers
• Helping reporters discover existing issues before submitting new ones

I’d be interested to hear if something like this has already been considered for BLT, or if it might align with the project’s roadmap.

If the maintainers think this could be useful, I’d also be happy to explore possible implementation approaches.