Last updated at Fri, 25 Aug 2023 19:21:23 GMT

Rapid7 is pleased to announce the release of Velociraptor version 0.6.5 – an advanced, open-source digital forensics and incident response (DFIR) tool that enhances visibility into your organization’s endpoints.  This release has been in development and testing for several months now, and we are excited to share its new features and improvements.

Table transformations

Velociraptor collections or hunts are usually post-processed or filtered in Notebooks. This allows users to refine and post-process the data in complex ways. For example, to view only the Velociraptor service from a hunt collecting all services (Windows.System.Services), one would click on the Notebook tab and modify the query by adding a WHERE statement.

Filtering rows with VQL

In our experience, this ability to quickly filter or sort a table is very common, and sometimes we don't really need the full power of VQL. In 0.6.5, we introduced table transformations — simple filtering/sorting operations on every table in the GUI.

Setting simple table transformations

Multi-lingual support

Velociraptor’s community of DFIR professionals is global! We have users from all over the world, and although most users are fluent in English, we wanted to acknowledge our truly international user base by adding internationalization into the GUI. You can now select from a number of popular languages. (Don’t see your language here? We would love additional contributions!)

Select from a number of popular languages

Here is a screenshot showing our German translations:

The Velociraptor interface in German

New interface themes

The 0.6.5 release expanded our previous offering of 3 themes into 7, with a selection of light and dark themes. We even have a retro feel ncurses theme that looks like a familiar terminal…

A stunning retro ‘ncurses’ theme

Error-handling in VQL

Velociraptor is simply a VQL engine – users write VQL artifacts and run these queries on the endpoint.

Previously, it was difficult to tell when VQL encountered an error. Sometimes a missing file is expected, and other times it means something went wrong. From Velociraptor’s point of view, as long as the VQL query ran successfully on the endpoint, the collection was a success. The VQL query can generate logs to provide more information, but the user had to actually look at the logs to determine if there was a problem.

For example, in a hunt parsing a file on the endpoints, it was difficult to tell which of the thousands of machines failed to parse a file. Previously, Velociraptor marked the collection as successful if the VQL query ran – even if it returned no rows because the file failed to parse.

In 0.6.5, there is a mechanism for VQL authors to convey more nuanced information to the user by way of error levels. The VQL log() function was expanded to take a level parameter. When the level is ERROR the collection will be marked as failed in the GUI.

A failed VQL query

Query Log messages have their own log level

Custom time zone support

Timestamps are a central part of most DFIR work. Although it is best practice to always work in UTC times, it can be a real pain to have to convert from UTC to local time in your head! Since Velociraptor always uses RFC3389 to represent times unambiguously but for human consumption, it is convenient to represent these times in different local times.

You can now select a more convenient time zone in the GUI by clicking your user preferences and setting the relevant timezone.

Selecting a custom timezone

The preferred time will be shown in most instances in the UI:

Time zone selection influences how times are shown

A new MUSL build target

On Linux Go binaries are mostly static but always link to Glibc, which is shipped with the Linux distribution. This means that traditionally Velociraptor had problems running on very old Linux machines (previous to Ubuntu 18.04). We used to build a more compatible version on an old Centos VM, but this was manual and did not support the latest Go compiler.

In 0.6.5, we added a new build target using MUSL – a lightweight Glibc replacement. The produced binary is completely static and should run on a much wider range of Linux versions. This is still considered experimental but should improve the experience on older Linux machines.

Try it out!

If you’re interested in the new features, take Velociraptor for a spin by downloading it from our release page. It’s available for free on GitHub under an open source license.

As always, please file bugs on the GitHub issue tracker or submit questions to our mailing list by emailing You can also chat with us directly on our Discord server.

Learn more about Velociraptor by visiting any of our web and social media channels below:

Additional reading:


Get the latest stories, expertise, and news about security today.