What is xarc?

XArc is a command line tool for browsing, searching, and exporting an X (formerly known as Twitter) data archive entirely on your own computer. It never uploads your data anywhere. Instead it builds a local SQLite database with full text search support, covering your posts, Note Tweets, and Articles, so you can look through your timeline, read the context around a post, and find any local media files that belong to it. The original archive files are never changed or copied. It runs on Linux or WSL with Python 3.11 or newer, and needs a build of SQLite that supports the FTS5 search extension. Native Windows is not supported yet because the tool relies on a Linux only file locking feature to keep the index safe and read only. There is no network access and no third party dependencies at runtime. To use it, you unpack your X archive, copy the example config file, point it at the archive's root folder, then run the build command to create the index and the stats command to see a summary. It can also be installed into a Python virtual environment so you can run it as a plain xarc command instead of a longer module path. The tool offers commands for listing accounts, searching post text or file names, browsing a filtered timeline by type, date, and order, viewing a single post with its thread and media, checking configuration and index health, and exporting search or timeline results to markdown, JSON, or CSV files. Multiple X accounts can be merged into one database or kept in separate ones, and most commands can be limited to a single account. Exports go to a dedicated folder by default and will not overwrite existing files unless told to. Search today is lexical only, matching exact words or literal substrings, with a semantic search mode planned but not yet built. The project also documents which private data, such as direct messages and account security information, is recorded only as file paths and never indexed or exported. It is released under the MIT license.