BurpSqueezer – Turn Burp Suite XML Dumps into Compact LLM-Ready Markdown (github.com)

BurpSqueezer is a newly released security research tool that transforms large Burp Suite XML HTTP traffic dumps into compact, structured Markdown representations designed for easier analysis by large language models (LLMs). By streamlining the data, BurpSqueezer alleviates the burden on LLMs to parse and filter vast amounts of raw traffic, improving efficiency in analyzing web applications. The tool significantly reduces the dataset size while maintaining crucial structural and relational information, leading to Markdown outputs that can be several hundred to over a thousand times smaller than the original XML dumps. This innovation is particularly significant for the AI/ML community as it enhances the ability to perform LLM-based analysis of web applications, especially under constraints like file size and context-window limitations. BurpSqueezer employs statistical and heuristic techniques to identify meaningful relationships within HTTP traffic while filtering out redundancy and low-value data. The tool operates universally, without hardcoded API patterns, making it adaptable for various applications with intricate business logic. However, it is essential to note that while BurpSqueezer streamlines data for analysis, it does not replace the need for manual security assessments and verification, serving as a complementary resource for security professionals and researchers.