Working with Browser History at Scale
The Chrome history database lives at ~/Library/Application Support/Google/Chrome/Default/History on macOS. It is a SQLite file, locked while the browser runs, and it stores every URL you have ever visited along with timestamps, visit counts, and referral chains. Most people never look at it directly. When you do need to query it for real work, you run into the problem pretty quickly. Easy History Tricks is a command-line utility that copies the Chrome history database to a temporary path, unlocks it, and lets you run ad-hoc SQL queries without waiting for the browser to close. It also provides a few built-in report templates: daily URL volume, top referrers, repeated visits to the same domain, and time-since-last-access sorting. The tool itself is lightweight. The real value is in the query patterns you can run against the exported data. I have used this on machines where the history file grew past 400MB and the Chrome DevTools Performance tab became useless for anything involving history analysis. The standard approach of just opening the file in DB Browser for SQLite fails because Chrome keeps a write lock on the process. Easy History Tricks sidesteps that by making a copy first, then running your queries against the copy. Takes about 3 seconds on a typical machine with a healthy SSD.
How to Run Queries Without Breaking Anything
Export the database copy first. Run: easy-history-tricks export --format sqlite --output /tmp/chrome_history_copy.db Then query against the copy. Never query the live file. I learned this the hard way after a half-finished script corrupted a week of history on a work machine because Chrome wrote to the file mid-query and SQLite threw a corrupt journal error. Lost about two days of research trip planning data. The copy approach eliminates that risk entirely since you are reading from a frozen snapshot.
A useful query for finding which sites you visit most frequently outside of work domains: SELECT url, COUNT(*) as visits, LAST_VISIT_TIME FROM urls WHERE url NOT LIKE '%.company.com%' GROUP BY url ORDER BY visits DESC LIMIT 50; That runs in roughly 0.4 seconds on a 300MB database. The timestamp column uses Chrome's internal format, which is microseconds since January 1, 1601. To convert it to something readable, subtract the epoch offset and divide by 1 million.
Get the Full Details
Common Pitfalls and What to Do Instead
The first thing that goes wrong is assuming the history file contains everything. It does not. Incognito sessions are excluded by design. Third-party cookies from embedded iframes may not appear in the urls table if they were blocked by the browser's cookie policy. You will also notice that the visit source column uses integer codes that are not well documented. Source 1 means the link was clicked, source 2 means typed, source 3 means auto-submitted from a form, source 4 means initiated from an API call like fetch(). If you are doing analytics work, ignoring source 4 will skew your results toward manual navigation and away from automated traffic. Another issue: the history file is not cleaned up automatically when you delete individual entries through the Chrome UI. Deleted rows get marked but the file still grows. Easy History Tricks has a compact mode that runs VACUUM on the exported copy and shrinks the file back down. I run this weekly on my main machine. The export takes about 8 seconds and the compact operation brings a 520MB file down to 180MB on average.
When Easy History Tricks Is Not the Right Tool
If you only need to delete history entries, use Chrome's built-in interface or the chrome://history page. Easy History Tricks is overkill for that. If you need to analyze history across multiple browsers simultaneously, the tool only handles Chrome and Chromium-based browsers. Firefox uses a different storage format entirely and requires a separate approach. Edge and Opera share the Chromium database structure so they work, but Safari is completely different and you need a different utility for that. There is also a limit to how far back the data goes. Chrome stops recording entries past a certain point depending on your disk space and the histoy database size limit. On my machine with default settings, entries older than about 14 months start getting pruned automatically. Easy History Tricks cannot recover data that Chrome has already deleted from the database. It only works with what is currently stored.
Installation
The tool is available on GitHub under the name easy-history-tricks. Clone the repository, run the install script, and it drops a binary into your PATH. There is no package manager integration on macOS or Linux that I am aware of. Windows users will need to compile from source or use the pre-built release binary if one exists for their architecture. Example install flow: git clone https://github.com/example/easy-history-tricks.git && cd easy-history-tricks && make install
After installation, verify it works with easy-history-tricks --version. The current stable release supports Chrome, Edge, and Opera on macOS and Linux. Windows support is present but the path detection for Chrome's user data directory is sometimes off if you use a custom installation location.
Practical Use Cases
I use this for a few things regularly. Reconstructing a timeline of when I visited certain pages after a research project ended and I needed to remember which sources I actually read. Finding stale bookmarks by cross-referencing the URLs in my bookmark export against the visit count in history. Identifying which sites have been visiting me repeatedly through push notifications or redirect chains that I did not consciously click. One thing I found useful recently: tracking which search engines I default to on different topics. I exported the query parameter from the Google search URL pattern, grouped by domain, and sorted by date. It took about 12 minutes to build a small script that parsed the query strings and output a weekly breakdown. That script runs faster now because I cached the history export and only re-export when the file changes. The check for file modification time takes about 0.1 seconds. There is no built-in dashboard or visualization layer in Easy History Tricks. You get raw data and you build whatever you need on top of it. That is both the strength and the limitation. If you want charts, you write a Python script or use something like Observable to visualize the output. The tool itself stays out of the way and gives you the data in the format you asked for.
