SimpleLogSearch is a free log file search that runs entirely in your browser. Click Upload, pick a log, type a query with AND, OR, NOT, and quoted phrases, and read the matching lines with their original line numbers. Nothing is sent to any server: your file is read into page memory with your browser's own file API, the search executes on your machine, and everything is discarded when you close the tab or load another file. There is no account, no signup, and nothing to install — it is a search box with real boolean syntax sitting on top of an ordinary text window. The query box accepts the same phrasing you would use in grep or a SIEM search bar, so existing muscle memory carries over. It was built for security work — access logs, event exports, application debug output — but it handles any line-based text. New to the syntax? The query syntax reference walks through every operator with examples.
The whole workflow is three steps, and there is nothing to install. A worked, step-by-step investigation walkthrough lives in the how-to guide.
"MID 123456" and "timeout" finds the lines that contain both strings, in any case.Six real queries and what each one finds. Every operator, with more examples, is explained in the query syntax guide.
"MID 123456" and "MID 123654" not "192.168.1.1" — the canonical example: lines mentioning both correlation IDs, minus the noise coming from one known host."status=200" OR "status=500" — every successful or failed HTTP response in a web access log, in one pass."svc=db" AND ("deadlock" OR "replication") — database-service lines that report either a deadlock or a replication event."192.168.1.1 " — the exact IP address; the trailing space keeps 192.168.1.10 and 192.168.1.100 out of the results."svc=api" "status=200" — implicit AND: lines containing both phrases, with no keyword required between them."and" — a quoted operator is just a literal word, so this finds lines that contain the text and.Short answers to the ten most common questions — the full FAQ carries fourteen, and the privacy policy states the data story in full.
No. Your log file never leaves your browser. The app is fully client-side: your browser reads the file into memory with its own file API, JavaScript running on your machine tokenizes and searches it, and everything is forgotten when you close the page or load another file. There is no upload step, no server to host your data, nothing in cookies or local storage except the interface language you choose, and no analytics reading file contents.
Up to 50 MB per file; anything larger is rejected with a clear error before it is read. Large files stay smooth because the results window is virtualized — only the rows currently in view exist in the page, so even a 10 MB log with 100,000 lines scrolls without stutter. Reading happens in the background while the status row shows the file name and a reading notice, then the total line count appears.
Yes — any line-based text file works. The engine treats each line as a plain string and applies case-insensitive substring matching, so it never needs to understand the format. Plain application logs, syslog output, web server access logs, Windows Event flat-text exports, CSV spreadsheets, JSON-lines data, and XML dumps all work as long as every record sits on its own line and the file is UTF-8 text. Binary files are not supported.
No. Matching is case-insensitive substring: error, Error, and ERROR all find the same lines, so you can type queries in whatever case feels natural. One subtlety comes with substring matching — there are no word boundaries, so searching for 192.168.1.1 also matches a line that contains only 192.168.1.10. The Query Syntax reference explains the trailing-space trick that narrows these collisions down.
Yes. The layout is responsive down to phone width — the search bar, buttons, and results window all adapt, and the interface is verified at 390 pixels wide. Searching large files works on a phone too, though a desktop screen makes long result sets more comfortable to read.
Because there is nothing to charge for: no server, no database, no per-seat licensing — the entire site is a few dollars of static hosting on GitHub Pages. To help cover even that modest cost, the site is monetized with Google AdSense advertisements, and that is the complete model. There is no premium tier, no paywall, and your files are never shared with any advertiser; the Privacy Policy page states all of it in full.
WebMCP is a W3C standard that lets a web page expose its capabilities to AI agents as tools. SimpleLogSearch registers four — get_log_status, search_logs, reset_view, and load_sample_log — so an agent can see what file is loaded, run queries, reset the view, or load one of the built-in samples. You do nothing special: in a WebMCP-capable browser you ask your assistant in natural language, for example to find every 500 status line except the health-check traffic, and it calls the tools for you.
Every result is labeled with its original line number from your uploaded file, starting at 1 — filtering never renumbers. A match labeled 4,182 is line 4,182 in the file: open it in an editor, SIEM, or any other text tool and jump straight there. Line counting treats the file as it is on disk, and a trailing newline at the end does not create a phantom extra line.
Reset only clears the search box and any error message — the file stays loaded and the window returns to the full log from line 1. Uploading a new file discards the old one completely: the previous data is thrown away, the search input is cleared, and the new file is shown from its first line. Since files live only in page memory, closing or reloading the page forgets everything as well.
Any evergreen browser: current versions of Chrome, Edge, Firefox, and Safari all run the full app — upload, boolean search, 50 MB files, and the resizable results window. One feature set is newer. The WebMCP agent tools listed in Features need a client with W3C WebMCP support; today that means Chrome 149 or later, Edge 150 or later, or ChatGPT Desktop. In other browsers the page loads as an ordinary website, silently and without console noise.
SimpleLogSearch is built and maintained by Chris Pardue — the about page covers why the tool exists and what it runs on.