Lishi 2-in-1 have been around for a long time, both for automotive and more recently for pin tumbler locks. Sadly, these tools are quite costly as they are keyway specific. Furthermore, this type of tool can’t be made for some locks as the keyways are too tight. Or so I thought, as of writing there are several 2-in-1 for sale for the paracentric Yale keyway. As I was intrigued, I’ve bought an off brand 2-in1 for CISA. In testing the tool, I’ve found various limitations that might impact the usefulness of this tool.
Let’s start from the beginning, Lishi is the brand name of a series of lockpicking tools designed and made by Zhi Qin Li. The Lishi company split up and Zhi Qin Li still sells his 2-in-1 under the brand Original Lishi, while another company sells them under the brand Genuine Lishi.
Original Lishi sells a variety of tools, the one generally referred to as a Lishi is a 2-in-1 lockpick that both applies a turning force and a tool for picking individual elements. The tool can also be used to decode the lock once the lock is open, and a key can be cut in the field with another of Li’s tools.
So, what does a Lishi 2-in-1 lockpick do? The tool consists of two parts; the body that is used as a turning tool and the thin feeler that’s used as a lockpick. This in itself would not be too useful, however, the body has a chart of where the lockpick is in the lock. You move the pointer to the desired element, indicated by the vertical lines, and push down lightly on the pointer. This in turn moves the lockpick, pushing down on the element in the lock. You can feel if the element is binding or not. And just as lockpicking, you go through the lock, pin by pin, and feel for the binders. Then you set each binder and search for the next one, until all elements are set. Click on one, click on three… Open!
From y2k these tools have been available for automotive locks as the combination of open keyways, many wafers, and typically low tolerances work very well for this tool. 2015 was the year 2-in-1 picks became available for pin tumbler lock for the USA market. (Schlage, Kwikset, and Master). Most of these locks have wide keyways, low tolerance, and very few security pins.
I’ve played with a few of these tools, but didn’t find them too useful. I’m not a locksmith, not in the USA. For me, they would be mostly a novelty. But the pick I’m about to show can be a game changer as it targets European locks I’m familiar with.
In December, I was notified a seller on AliExpress sells 2-in-1 (not a Lishi!) for the paracentric Yale keyway. The consensus under lockpickers is that this tool could not exist, because the keyway is too tight and has no straight access to the pins. I was curious enough to fork over €50, and bought one for CISA as it’s very close if not identical to the Abus C83, the lock we use for impressioning championships.
The seller is quite open about the tool’s limitations and wrote on the lever “80% coverage. Without pin 8 or 9”. While this sounds like it’ll open 80% of locks, but it doesn’t seem to be the whole story, as we will find out. The biggest concern with a tool that works on a subsection of locks is if the user can detect the tool does not work, instead of user error or lack of skill. I suspect so, but it will be far from easy. In short, a lock will not work with one or more cuts deeper than a 7 and therefore this pin will always be overset and this you can detect.
As I was curious about the 80% claim, I’ve spent an evening measuring my Abus C83 keys. While these are not CISA, they are close, and I happen to have a modest collection of these keys. 92 out of 283 of the measured keys have no cuts deeper than 5.5mm, the size of the tool. This means the tool will only work on 30% of my Abus C83. This is consistent with a statistics sanity check. For this, we assume every lock has a uniform distribution of cuts, ignoring MACS. This came to be (7/9)^5 = 28.4%.
I’m considering this 30% an upper bound, as Abus C83 and CISA aren’t shipped with standard pins. The old locks are shipped with mostly spool pins, and the new ones have serrated, spool, and T-pin key pins and the same for the drivers.
Besides the theoretical usability and security pins, what other flaws would make this tool suboptimal? The picking tip snags while moving from pin to pin. Furthermore, picking in the counterclockwise direction binds the picking tip, and it makes it difficult to differentiate between a binding pin and a binding picking tip.
Let’s wrap it up, this tool is sold at €50 and promises quite a lot. However, theoretical, it will only open 30% of all the locks it was designed for. Furthermore, the limitations of security pins and rotation direction will limit the functionality even more. A practiced lockpicker might be-able to overcome some limitations, or detect the tool will not work. A tool that only opens a very small subset of locks is not a very useful tool, and I can’t recommend it to pick these locks. However, as new pickers always struggle to find the binders, and this tool enables them to actually ‘see’ what they are doing, it could be a game changer for teaching.
Pictures CCBY4.0 Jan-Willem Toool Blackbag
Great review. The issue with being unable to open most of the locks is why I never went further with this type of pick design for European locks. I just bought an SC1 model to try (as briefly discussed) and now I have to try to find a Schlage to test it!
I do expect this tool can be made, just not as easily. The best thing would be for the lockpick to rotate around the warding. Time will tell if it can be done.
As far as I know, The SC Lishi works great. I’ve picked an SC with standard pins and a borrowed Lishi, at the time I didn’t find it easy, and it took me a few hours to grasp the technique.
Do ask for the locks on Discord, as there must be loads in the UK already, if not I’ll send you some.
Nice article. I’ve never used these tools, but they are fascinating and it’s interesting to see how tight they can push the keyway. A nitpick on terminology: “normal distribution of cuts” should be “uniform distribution”. Normal would mean a bell curve, where 4/5/6 were common and 0/1/8/9 were very rare. Uniform means all are equally likely.
You are totally correct. It is a marvel they can make such a tool, but of course I would have liked it to work in more of my locks. I’ve corrected the remark about pin distribution 🙂