Angle-grinder is a command-line log analytics tool that lets you parse, filter, and aggregate log data with live-updating terminal output.
The tool solves the problem of analyzing logs when you lack access to dedicated log aggregation platforms. It works by accepting a query language composed of filters and operators that transform and analyze streaming log data. You can parse JSON, logfmt, and delimited formats, then apply aggregations like counting, summing, averaging, percentiles, and sorting. Results update live in the terminal as data is processed, and the tool can handle over one million rows per second for simple pipelines.
Angle-grinder suits developers and operators who need ad-hoc log analysis on local machines or servers without centralized logging infrastructure. It works well for one-off investigations, performance analysis, and situations where you want to avoid the overhead of setting up external services. The tool is particularly useful when you have access to raw logs but need to perform sophisticated analytics quickly. It is designed as a lightweight alternative to platforms like Splunk, Honeycomb, or Kibana when those are unavailable or impractical.
The project shows active maintenance with binaries available for Linux and macOS, and support for additional platforms through source compilation. The tool includes a self-update mechanism and is maintained across multiple package managers. The codebase is written in Rust and supports a functional programming language for query composition, with features like field escaping, complex filtering with boolean operators, and user-defined aliases for common analysis patterns.
