A few days ago, an email came out of the blue, demanding that I install an "agent" from a company named "Drata"* on my laptop. The motivation is that my client badly want a SOC 2 certification.
I have worked as a developer for more than 30 years. Tiny shops. Startups. Major league. I have never even heard about someone putting agents on developers laptops.
I'm pretty pissed off. So are the teams I work with.
Is this the new normal now?
Just for the record: I don't have credentials to production systems, and I don't work with production data. I just figure out how to transform dreams into code, I write parts of that code, and then I fix it as needed.
* Drata (https://drata.com/about) is on a "Mission to Help Build Trust Across the Internet". Their business model (in my case) seems to be to take money from companies to spy on their employees/contractors, and then they sell the employees/contractors private information to "targeted advertising". When I confronted them about this, they replied: "Feel free to reach out to your Drata administrator internally with concerns. Do note, that when your company contracted with Drata, any edits or redlines they provided will prevail for all employees of your company." - basically to just bend over and smile.
I talked to an Intel HR person (informal chat, I never applied there nor planned to) 2-3 months ago and after I stated that after a decade of remote work, I see pandemic-driven introduction of harmful concepts like spying on previously trustworthy contractors by control-freaking managers that have no idea how to prove themselves in new reality, I was given a look which would usually be reserved for a psychotic person believing that they're being watched 24/7. Quite an unique experience, contrasting with how HR folks are trained to do sect-like "love bombing".
You want me to work for you and deliver results? My pleasure - that's what I do.
You want me to hang a company logo in my place and sit in front of camera multiple times a day, log every minute of my time and creep on me in other ways? I never worked in "Office Space"-like environment and I'm not planning to. Go fuck yourself, I'm out.
* Drata is a vendor that helps a company navigate your SOC2 compliance process, by organizing all the controls and helping you gather evidence that you have done so. For instance, they'll connect with Github and make sure everyone with access to your repos is a company employee. If you don't use Drata you have to gather this evidence yourself, repeatedly over months, and it's a pain.
* The Drata agent is a pretty innocuous thing. It checks you have done things like turn on disk encryption, have updates enabled, and that the screen locks if you walk away. It does NOT monitor employee's activities. These kind of security checks are incredibly common and are required for certifications like ISO27001 and SOC2. SOC2 is not really optional for large enough b2b SaaS.
* The poster says "Their business model (in my case) seems to be to take money from companies to spy on their employees/contractors, and then they sell the employees/contractors private information to "targeted advertising".
Do you have any evidence for this?? I've just been involved in selecting Drata as a vendor for SOC2 compliance planning for our company. If this is true it's a huge deal and totally against my understanding of their business model. It honestly sounds like bullshit to me! But if you have evidence that they do this, please let us know.
* As a freelancer, whether you are required to install security monitoring software is definitely an open question. If you're delivering work separately and not connected to company systems, then ok. If you're basically just acting like any other employee, and connected to the company systems, then you will probably have to do this. Because otherwise they would fail SOC2 and managing your legal status as "Freelancer" vs "Employee" (for tax reasons??) is not worth not being certified.
It sounds like the spyware is non negotiable and I personally wouldn’t have issue with it if the client also provides a laptop on which to run it. The client is free to do whatever the hell they like with their own hardware.
What’s objectionable about situations like this is the client wanting to have their spyware cake and eat it on someone else’s computer. That’s a great deal for them — why pay $5k for a laptop when you can just pay $100 for the spyware license instead?!
I realize the economics aren’t exactly on point, but I tend to view situations like this as them stealing $5000 — my laptop — from me.
So moving forward:
1/ you are “happy to help them reach their compliance goal and move to a Drata controlled environment”
2/ to do so you “will need to isolate them, as a client, to an airgapped environment solely for their work”
3/ which will need “$5k up front and a lead time of a week to order new hardware, or for the client to ship you a preconfigured laptop with configuration X Y and Z.”
Saying yes with principled conditions is always a good route forward. Yes-but instead of no-but.
Installing agents is a sign that the company doesn't value trust with their employees and treats them as liabilities. Companies that made me an offer who had asked me if I would have an issue installing an agent just got rejected from my employer list. If they don't trust me doing my job why should I trust them doing their job. Why not install an agent to the CEO's computer as well? After all I should trust him that he's doing his job well enough for us not to lose our jobs. I'm also dependant on him after all. These are all relationships where trust plays a major role.
After all, if you think an employee isn't performing you can just have an annual PDR (Performance and Development Review) and figure out if you have to get rid of that employee or talk to him. Why spy on all of employees? Agents are just an excuse, not the means. It's a disgusting in my opinion excuse to spy on everyone.
Whatever the case or how common this is, I won't ever accept agents to spy on me. You do you but I think everyone should do the same. I demand your respect to be mutual to my and your privacy and sense of trust. Agents are harming the remote development space and skew the perception of what it means to have a healthy team.
Not only is your personal equipment yours, it likely contains information about other businesses that you probably have legal obligations to, like NDAs, and if the client doesn't understand that you can't ethically violate that for them, then they're not a client you need to be dealing with.
It doesn't help that the 3-5 "IT agents" they run are rarely doing anything useful expect fulfilling 3-5 different directors idea of spying.
It’s an interesting question, and one that we’ve evaluated with respect to our own customers, who lean on freelancers and other small vendors. Getting back to my original statement about the nature of the business relationship, we’re asking questions about what level of technical sophistication a freelancer has and whether they’ve established enough of their own policies and procedures to meet regulatory requirements independently.
Often times, that’s just not practical. Even then, my preference is for the client to provide a managed device to the freelancer or to offer them self-managed options with documentation requirements to prove compliance. Forcing a contractor to install an agent like you’re describing onto their own device feels like a privacy intrusion, and may also represent a risk to the contractor’s other customers.
Then they came with the same shit for our phones, but since everyone plainly refused, they found another solution for that. Because there always is one. But this was nipped in the bud at the start. Had people accepted it, there would be no going back, because there never is going back.
At one point, I was writing a small demo in golang for one of our projects and I've been contacted by a security engineer telling me that I've been hitting C:\Users\ Needless to say, this dragged on for a week due to complex internal politics. I thoroughly enjoyed a week of paid time off.
If you don’t want to, make the pain to them apparent as a line item. Tell them complying with this is increasing your costs (by the need to protect other clients’ information) and that your rate will increase by Make them see the costs of their decisions. We had timesheets instituted for a prior company. I had all my devs add an explicit time for entering timesheet data under a dedicated project code. 15 minutes a week for every dev plus 30 minutes for every lead all billed to one project adds up pretty quickly. "Why are we spending 1% of our time on filling out timesheets?!" "That is an excellent question."
Really weird stuff, I hope that won’t become a trend.
The customer provided me an image with all necessary licensed software I need to use to provide the service, including software to connect to their production infra via VPN and encrypted filesystem. They don't have access to camera, mic, my private system or LAN (use NAT mode). Clipboard sharing is set with guest to host direction only. I can filter out any call home stuff I don't like on my router unless it goes thru their VPN. I Zoom them with my host system, then do screen-sharing of virtual machine window only.
In my opinion separating customer gigs with VMs in general is a safe way to prevent accidental cross-customer data leak. Of course it depends on kind of work you do and software you use, however personal licenses often allow to use software on multiple devices by the same person.
I don’t conduct personal business of any sort on a corporate device. Just not having direct access to production won’t exclude you from security protocols, else how can you guarantee nobody slipped adjustments into software you have checked out,a ‘git push’ originating from your endpoint, which then gets deployed?
Is it so difficult to define deliverables and pay for completed and tested work? The business value of any given function point is the same whether it took 100 hours to develop or 10 hours. Of course, more productive programmers would benefit under such an arrangement.
Oh wait ... the problem is that requirements specifications are never clear nor complete enough and there aren't any tests to confirm correctness of the implementation.
That said, as I work from home, my work laptop lid remains closed for all but a fortnightly company all-hands meeting, and I ensure that I keep zero personal data on it. I'd be an absolute no if the demand ever morphed to always on video or activity trackers. That's a bridge too far.
As it stands, I understand the need for some policy enforcement/remote control of their assets, but will make whatever moves I must to ensure that policy doesn't infringe on the rest of my environment.
Do we need an open-source repo/db/blockchain to track companies which track workers? This could monitor certification/regulatory requirements, benchmark tracking across near-peer companies, real-world impact on human performance, and supply chain integrity of tracking vendors. If a tracking company is breached, it would be possible to flag all of the companies using that vendor.
kvm switch are great to be able to re-use your input/output devices without mixing up personal and corporate stuff... this is what I use... A physical button to separate the 2 environments.
This is also part of the reason why BYOD is going nowhere as the second the company wants to audit/control setting it's no longer "Your Own Device".
The screenshots can then be as part of the dispute resolution process, and can also protect the contractor in case of disputes from unscrupulous companies.
The agent captures screenshots every few minutes, and the contractor can review and redact any screenshots before sending them.
Adding a monitoring agent for an existing contractor is a major change in the contract terms, and not something that I would consider acceptable.
It's not normal, and not something you should agree to.
If it's your own personal laptop, then it's your property, they have no right to make you install something on it. If it's their laptop they've given you, then they do. And it's up to you whether you want to work in that way. But as I said, that would sound alarm bells for me.
Not sure what the other commenters in this thread are going on about but AICPA's soc2 common criteria _do_ require that a bunch of that stuff is configured. The reality we're facing is that unless we actually monitor for those basic security config things, sales/marketing/etc will disable those setting for no reason and promptly leave their laptop in a Starbucks with client user lists or confidential data on it.
For other context - based on our research, compliance automation platforms like drata or secureframe greatly decrease the cost of the actual audit since it makes evidence collection that the proper security controls are in place and are functioning much easier.
From your perspective though I 100% get the concern though from a freelancer - I'd say that they shouldn't want you to be handling their source code on your personal machine anyways and should prob. send you a laptop.
Big firms when they issue their own hardware often install such agents that have ability to not just monitor activity, but also wipe out data or change user account credentials.
I personally find it weird if the ask is to install such agent on BYOD (personal device), since not just the company data, but your personal data can also be wiped out remotely, or your account credentials can be changed remotely locking you out of your own device.
You can choose not to install it and upload a couple screenshots of requested settings (having disk encryption enabled, having a password manager installed etc) periodically. If the client forces it that is a little unreasonable because drata tracks if you uploaded stuff in a timely manner anyway.
They don’t need to force it.
If it's your own laptop, you don't have to do that. That's yours, not their property. They can provide you with one configured as they request, or provide it and ask you to configure it that way. Or you could set up a VM for it.
Corporate spyware is kinda common nowadays. It's mildly annoying but mostly unlikely to be a problem in many/most places. Mostly just there to deal with problem situations.
And you mention that your client wants SOC 2 certification - chances are they'll never actually bother hiring someone to watch what you do on your computer, they just want to be able to check off a box on a form that says "yeah we do this, and all our employees have this thing installed, so we have central control of our data." to get the certification. Because that's what it's about. But also it's just bureaucracy, and probably just checking that box is all they care about so they can tell their clients/customers that their solution is officially certified safe. A lot o' stuff like that is driven by, and ultimately, just feature checklists.
I spent about 15 years consulting, on short- to medium-term often-recurring projects. Most clients didn't ask vendors to instrument their machines. Some did; for those clients, the solution tended to be that the client provided us with a machine to work on.
Most of these agents are truly awful. I don't know anything about Drata. You should not be psyched to have that running on your machine; I would isolate it somehow so that it's only in contact with that one client's workload.
But they're not making up the SOC2 thing. It's pretty likely they won't budge from this, not because they really care about the agent thing, but because they really do have a documented SOC2 process with "agents on desktops" as a stated control (almost everybody with SOC2 has some kind of agent somewhere, though you usually hope it's just MDM). They do not have a choice about whether to tool you up; your choice is likely just to stop working for them or not.
That said, our usual approach to dealing with customer-required installs like VPN clients is to just spin up a VM using VMWare Workstation on our development machines and do all of the things that touch their network with that. Given the nature of our work, we connect to their environments as little as possible and we leave those VMs off at all other times. We haven't had any problems with that approach thus far.
Additionally, we don't offer our clients the option of giving us development laptops for our work with them. That just makes us churn hours without producing anything while we deal with whatever local IT silliness they have.
Technical considerations aside, the idea that they want to spy on their contractors is troubling and I'd get away from that situation as soon as possible. Unless they decide to pull back on these requests, it sounds like they'll be just be emboldened to micromanage even more.
Disk encryption, screen time outs, remote wipe etc. contractor machines with code and production access are treated as critical assets and are fully under IT control.
If this is really just your employer (only client), but you have a "freelance" relationship for tax purposes or whatever, then you might want to consider whether you will be better off just getting a job.
As a software engineer, I assume you have employment options (there is demand everywhere it seems), so you can probably afford to do what makes you happy.
* completely open source
* have gone through security audits with public reports, and a favorable outcome
* have reproducable and verifiable builds, and those are the only ones distributed, and the end user can easily verify that their binary copy is an official build?
Right?
Because if not, aren't you just adding another attack vector onto all your employee/contractor laptops when you use 'Drata' to check a policy box on your SOC2 application?
[edit: formatting bullet list]
As with other backdoors, these will leak important data and ultimately become priority attack vectors to steal or corrupt data.
Of course there's also the worker privacy, but that will always get trampled on until the workers revolt.
Part of selling yourself (whether your mind or your body) is deciding where to draw lines. What will you do, or accept, to get paid?
I won't accept invasive monitoring. Companies like this can look elsewhere (and they'll find people who will happily trade everything for a little money).
Another alternative might be to install it into a VM or old but freshly-paved computer.
There was one company which required devices to be up to date on the latest security updates from the OS and every wednesday an employee was chasing everyone to get confirmation that our systems were updated.
If a client would require an agent to be installed I would ask for a company laptop to do the work on.
Since the certificate is about protecting user data - and you say you do not have user data - then I would not just accept it, without trying to reason with them, that the general approach they are doing is maybe too broad and unneccesary.
If I was in your situation I would ask them if they want to discuss the arrangement, the value and the guarantees they are getting. I would suggest we can agree on zero notice period and no questions asked termination policy (of course symmetrically). I would also want to discuss how they will know the work progresses so that they are satisfied they are not being robbed. If that wouldn't work I would part our ways and find other job.
As to installing spying software that should be absolutely out of question. If you agree, you are just enabling them to do the same for other people.
Listen, there is no value in having spying software on your computer. Will you work more diligently when you know you are observed? That only works for menial jobs, but if your job is to do anything complex you are just burning time for no reason.
#2 - if they require something specific in terms of hardware ("install this spyware directly on bare metal"), have the discussion about hardware setup cost/time and then expense them for the hardware you have to purchase.
In my last four engagements, every single laptop provided by the employer had something. Usually Tanium or Carbon Black. Network interfaces being disabled entirely if you're not connected to their VPN. One client requiring the use of a Meraki hardware VPN appliance.
This was an investment bank, a university, a software company and a health insurance company.
It's normal to have software that safeguards company's intellectual property (how suitable this particular software is, I cannot speak). However, with that goal in mind, it's also normal for company to provide a dedicated, company-owned-and-managed hardware as well, such as laptop or phone.
Demanding that employee or contractor use personal hardware but install monitoring software seems a lose-lose proposition for everybody - it will not necessarily achieve the level of control and safeguards that company desires, and it compromises the contractor's ability to safeguard their own and other clients' data.
Depending on circumstances, my own approach would be to start with a friendly email indicating that you understand and support their goals, and propose that the best way to achieve them is to use customer provided and managed hardware.
The reality is though, after i had completed the intial client-app for the Pc's I called the client and terminated the project. He was not to happy about it and a lengthy discussion about whats right or wrong about it ensured. We agreed that it is really not neccesairy and also unwanted surveillance. so i was happy and the code has been scrapped.
....Until his inhouse staff taught him how to read the logins from the actice directory....
Truth be told, if they supply the hardware and you consent to it( at lesat in europe), they are within their rights. If you think its right and give your consent to it, is up to you though
PS: A permanently running agent is most likely to make screencaps too.
In that regard, its easy for me to put an agent on their vm, I do run pi-hole so it doesn't matter what VM I'm in, most of this sort traffic gets filtered.
I would also recommend reviewing the contract you have with them, to see if it allows them to put these sort of measures on you. And personally determine if this contract is worth keeping, a company wishing to push something like this, with that rigid a response, doesn't sound like someone you would want to maintain a relationship with.
But either way, it seems like this is something that could be resolved without much effort.
- It's a massive breach of trust (I'd consider just asking for that tool a testament of no trust at all, irrepairable actually)
- The job market gives them zero bargaining power
If it comes "out of the blue" like you say, chances are it's being driven by a new guy. You'd do yourself and your client a huge favor by immediately, and visibly to all stakeholders, pointing out the idiocy of that guy's idea (remember, the trust is gone already, no need to sugar coat it then).
Because there will be more of those ideas, if he's not interrupted, possibly harming the company in cataclysmic proportions down the road.
As a contractor, you are not an employee so not covered.
Basically you get to choose what to do, and in my experience this is not normal, although companies often do have IT requirements for systems that will have access to sensitive information, so the concept in general is not unusual.
For me, the fact that this isn't purely about security (e.g. it's not some agent that comes from Cisco or some legit vendor only interested in security), I'd say no. But it depends how hungry you are for work. Since software developers are hard to find, I'd expect you can find work from other clients that don't have this requirement.
Or explain them that I value privacy and if they don't, they can go search for another collaborator.
Or tell them that installing spyware on my computer is going to cost them 2x the money.
I love the PR. Why is SOC2 important. Because relentless unrestricted spying allows you to foster a control system like never before over your pathetic serfs that dare wish to maintain a work life balance. When that project you forgot to assign goes over deadline use SOC2 to flay them with human resources over too many seconds of bathroom time causing the project to fall behind.
When I’m not using my laptop it’s closed, so the camera is off and the mic is muffled.
It doesn’t seem like that big of a deal to me. The laptop is for work stuff, most of which is in the cloud under their control anyways. Worst they’re going to find is my raw and unvarnished work logs, which might hurt some feelings if anyone is over-sensitive.
One of them is reporting all the processes I'm running. Certain keywords will trigger IT to reach out to investigate.
Another of them is intercepting all my web traffic, even going as far as installing its own CA and decrypting SSL. It's fun when that hiccups and I start getting SEC_ERROR_REUSED_ISSUER_AND_SERIAL errors everywhere.
This provides great incentive for me to keep personal usage off my work computer.
The details come down to what exactly the software is spying on (I won't look at their website): if it's too invasive, even on a company-provided computer (as many other commenter suggested), it can run afoul of the GDPR (i.e. illegal). There have been some cases already on that [1], and looking up some of them might prove to be a very effective negotiating strategy against that.
[1] quick search on Google: https://www.complianceweek.com/data-privacy/employee-monitor...
Another solution: Virtualize a pc, put everything that you need to work for that client in that VPC, install they drata only in the vpc. Or better yet, ask them to provide you with a VPC image with what they want.
I can understand being piss about it, but unless you are ready to drop them over this, getting angry will be of no consequence
We do two things.
One: we give them the choice. Yes it costs money, but not that much.
Two: we went with Kolide. To understand how they are different, go read https://honest.security.
I've been a contractor on and off for almost 20 years, and no way would I let someone insist on an agent. The most they can ask for is a VPN client.
The soc2 controls in question can be met with a handful of config changes and an antivirus install. (I’ve implemented soc2 controls five times) Full disk encryption with FileVault or Bitlocker Screen lock Enable automatic security updates Use of password manager Virus scan with ClamAv or windows defender.
If they want an agent to make sure you’re working when you say you’re working I’d pass.
A) they're being super lazy and using Drata as a blanket to cover a bunch of SoC2 requirements
B) it has nothing to do with SoC2, it's just their excuse to push this employee spyware
Try asking for a company provided computer. I'd also put that company device on it's own VLAN.
Another option would be you set up something like that for working with that client and install their agent on it.
In the rare event I have a customer that let's me work on my own hardware, I use VMs. VMs make it easy to segregate and backup work.
You can then either build the cost into your rates or suggest you bill them for it. You could even sell the laptop when the project ends.
Of course you can also be principled over it if you don’t need the work. It is after all a B2B relationship.
They're not a client if they can make you do this, they're your boss. So don't let them make you do it, because you're not their employee.
If a company sends me decidated hardware, including an LTE or 5G modem to connect to the internet, they can install on that whatever they want.
Their hardware, their rules.
My hardware, my rules.
It is business.
Doing work for clients often involves installing other software of theirs i.e. their preferred VPN client. I just run up a VM because having 10 different VPN setups is a disaster waiting to happen.
Three of the last four (marketing) agencies I worked for supplied me with a MBP.
There are so many accurate responses in this thread. Like many have mentioned, SOC 2 is indeed not a prescriptive framework. Much of the confusion behind SOC 2 stems from that fact. It allows you to customize your InfoSec program to your company's needs. As we know, this can vary from company to company, hence why I read so many correct ways of approaching this specific situation in the thread.
Why SOC 2? SOC 2 is primarily customer-driven (this is why it becomes so urgent on your org). Buyer's require their vendors to undergo these third-party audits for their own vendor security management. While they would love to take you at your word, they feel a bit better knowing that a third-party took a look under the hood of your InfoSec program.
Employee vs. Contractor The legal status of an employee vs. contractor doesn't really matter for SOC 2 or most other InfoSec frameworks. At a minimum, what they really care about is the individuals ability to access, modify, view or otherwise have an effect on production/customer data. If an individual has that ability, they are likely in-scope (this can mean a lot of things). If an individual is indeed in-scope for your audit, they should follow your InfoSec program. You can always have carveouts for certain scenarios (for example, background checks are illegal in many countries so you may exclude them for individuals in those countries).
Company Policy What this all comes down to is the policy that the company has put in place. Does the company require all employees and contractors regardless of access to have hard drives encrypted without any carveouts? If so, then the company must follow that practice, or they will risk get an exception on their SOC 2 audit report. SOC 2 has some minimum standards that auditors look for but ultimately the company sets its controls and policies (if they are barebones they might not get accepted). Auditors are human and since SOC 2 is not prescriptive, reasonable minds will differ as to what those minimums exactly are.
Common Recommendation This has been mentioned a number of times in this thread but what we typically see and recommend is that you treat all employees as in-scope (this makes it easier on the company so they don't have to make determinations about who should and shouldn't be in-scope) and then for all contractors, you create a carveout where if they don't have access etc to production/customer data then they are not in-scope. In this case, such contractors would not need to track things like hard drive encryption, rendering the need for the agent moot. This seems in-line with the original posters role, and we would typically not have our customers require this of such a contractor.
There is nuance needed to make some of these determinations. For example, a company could hire a contractor who only has access to source code. In this case, an auditor may say that this contractor is indeed in-scope since they have control to modify source code that is pushed to production, even though they don't have direct access to the production itself.
We can't speak to the Drata agent, but based on what we would expect, the organization in OP's question is most likely trying to simplify evidence gathering when it comes time for the audit. There are other ways to grab such evidence (manual screenshots), but they are time consuming. Based on OP's job description it doesn't seem like its necessary for OP to be in-scope in this scenario and therefore the organization shouldn't need to collect such data. However, as we mentioned, this organization could have more stringent policies and without more information there isn't a wrong or right answer here. What we can confidently say is that it isn't a hard SOC 2 requirement.
This is especially true when you're connecting to their network and using their data. I'm sure there are exceptions for certain types of contracts, but for hourly work this is the norm for companies that take their data security seriously.
Full stop.
If your employee doesn't fulfill some other criteria, like providing all hardware containing such a agent, not requiring you to use our have the hardware powered on outside of you work time etc. it's unlikely to be legal.
They might also need to provide you with a way to separate internet access, e.g. an LTE modem, a this agent the to scan the network which they are in, which is also in beach of law in case of home office.
While I don’t know all the details about this specific case, I want to clearly lay out and address some of the concerns and misinformation in your post. We believe trust is the most important aspect of any business, and it’s why we’ve ALWAYS made a point to be very transparent. As a company, we would never develop any software that does what you are claiming. In fact, Drata does just the opposite, by helping companies protect data.
First, we should address the WHY. In order to be SOC 2 compliant, one thing businesses need to do is ensure their employees’ and contractors’ computers are configured securely. The “agent” is one way Drata efficiently does this, especially for teams with remote employees.
Now, let’s look at the HOW:
- The Drata agent is a lightweight, read-only osquery based agent that reads system information such as hard-drive encryption, screen lock timeout, firewall status, etc. Drata collects that information to ensure companies are meeting their security/compliance requirements and so that these companies can prove their SOC2 obligations are being met during audits.
- Regarding the TOS, there is no monitoring or selling of your personal information. That is unequivocally false.
The question you asked Drata about over email was directly related to the TOS, and not the agent (though we’ve explained what the agent does above). As we stated, we don’t sell customer data. We never have and we never will.
I would be happy to chat more to address any concerns you have. You can reach out to me at adam [at] drata.com to chat anytime.
This is neither normal nor reasonable. Do not accept it.
AFAIK it only checks a few items. We had to install it, too for SOC2 compliance. I installed it, let it send in a report (because I'm on Linux, I manually had to make a few screenshots as proof), then uninstalled it.
I'm happy to do the above process next time they complain, if ever.
Sometimes I really wish for a trade union of IT workers, however bad that sounds.
My response was that I'm contractually forbidden to allow that since on my device there's data GDPR-relevant and otherwise of other customers.
You can also make up a company-policy (even if you're a single person) and communicate it that way. "There's a strict company policy preventing me/us from installing this kind of software."
Apart from that, this violates the no-asshole-rule.
2) Consider getting a special laptop for them alone (and bill it to them)
3) Have them send you laptop for their work.
4) Go to 1
Specifically, it's perfectly reasonable for you to say "OK -- if you're willing to provide me with a dedicated laptop". They can say no of course, but so can you. Or you can request a rate increase (which they would probably say no to, if they won't provide you with a laptop).
Either way, those are you choices. Yes it sucks to a degree, but that's what work does generally and which is why it pays money. All we can do is moderate the suckiness-to-money ratio as best we can.