Skip to content
Notifications
Clear all

Thoughts on the new 'Code Security' module? Worth adding to our existing SCA tool?

20 Posts
19 Users
0 Reactions
247 Views
(@danielj)
Reputable Member
Joined: 4 months ago
Posts: 254
Topic starter   [#23782]

Hey everyone! 👋 We've been using Lacework's core cloud security features for about a year now, and the platform's been solid for our AWS workloads. Just got the email about their new 'Code Security' module, which seems to bundle SAST, secret detection, and infrastructure-as-code scanning.

We already have a dedicated SCA tool (we use Snyk) that handles our dependency scanning pretty well. I'm trying to figure out if adding this Lacework module would be complementary, or if it would just create overlap and noise. The main appeal for us is having one fewer dashboard to check, and potentially tighter integration with the cloud alerts we already get from them.

Has anyone piloted this new module yet? I'm especially curious about:
* How the secret detection compares to something like GitGuardian or TruffleHog.
* Whether the IaC scanning (for Terraform, CloudFormation) catches misconfigurations earlier than their regular cloud resource alerts.
* If the pricing is additive or if they have bundles for existing customers.

I'm leaning towards setting up a trial, but would love to hear from the community before I go building another comparison spreadsheet! 😄

— Dan


spreadsheet ninja


   
Quote
(@eliotk)
Estimable Member
Joined: 4 months ago
Posts: 111
 

Interesting question. I haven't tried the new module yet, but I'm in a similar spot. We also use Snyk for SCA.

My main hesitation is exactly the overlap you mentioned. Adding another tool for secrets and IaC might just mean more noise, unless the integration is truly seamless. The dashboard consolidation is a big plus, though. Have you seen how the findings from Lacework would be prioritized alongside your cloud alerts? That could be the deciding factor for us.

If you do the trial, I'd be curious to hear how the secret detection compares in practice.



   
ReplyQuote
(@annas)
Honorable Member
Joined: 4 months ago
Posts: 542
 

We're about three months into the pilot, and it's squarely in the "complementary" column for us, but with a major caveat.

The secret detection is on par, maybe even a bit noisier than GitGuardian, but the consolidation is where it wins for us. Having those IaC findings and secrets alerts roll up into the same policy engine as our cloud runtime violations means we can finally build unified suppression rules. We blocked a deployment last week because the pipeline IaC scan caught a security group rule their cloud scanner would have flagged two hours later. That's the integration payoff.

On pricing, push them hard. They offered us a 20% discount for adding it to our existing enterprise agreement, but it was still additive. Your comparison spreadsheet is still necessary. The real question is whether closing that loop between code and runtime is worth the premium over your already-working SCA tool. For our compliance requirements, it was.



   
ReplyQuote
(@charlotte4)
Estimable Member
Joined: 4 months ago
Posts: 99
 

That's a really useful detail about the unified policy engine. I hadn't considered the suppression rules angle. It sounds like the value isn't just one dashboard, but a single rule set.

When you say you blocked the deployment with the IaC scan, was that an automated gate in your pipeline, or a manual review triggered by the finding?



   
ReplyQuote
 dant
(@dant)
Honorable Member
Joined: 4 months ago
Posts: 434
 

> the value isn't just one dashboard, but a single rule set.

Exactly, that's the architectural distinction that matters. The unified policy engine is the key integration point, not the UI consolidation.

In our case, the IaC block was an automated gate. We configured the Lacework agent in the CI pipeline to fail on specific, high-severity IaC violations, like overly permissive security groups. The policy triggering that failure is the same one that generates our runtime cloud alerts, so there's no drift between pre-deployment and post-deployment rules.

A caveat, though: achieving this requires mapping your SCA tool's findings into that same policy framework, which isn't trivial. If Snyk's vulnerability data lives outside Lacework's engine, you still have a split-brain scenario for dependency risks, undermining the single rule set goal.



   
ReplyQuote
(@finops_auditor_ray)
Honorable Member
Joined: 8 months ago
Posts: 467
 

The dashboard consolidation is a trap. It's just moving pixels around. The real cost is the engineering hours spent tuning out false positives from a new engine.

You mentioned pricing. It's always additive. They'll show you a bundled rate, but run the math on your current per-container cost and see where the new module lands. If they won't show you the line items on a sample bill, walk away.

Before you even consider the trial, get them to prove the IaC scanning catches issues *earlier* in your actual pipeline, not just in a demo. Otherwise you're paying twice for the same finding.


show me the bill


   
ReplyQuote
(@helenr)
Honorable Member
Joined: 4 months ago
Posts: 534
 

That's a smart set of questions to take into a trial. Based on what others have mentioned about the unified policy engine, I'd suggest you specifically test for that 'split-brain' scenario user1330 hinted at.

For your third question on pricing, I've generally seen these modules priced additively, but discounts for existing enterprise agreements are common. The key is to negotiate based on the specific resource coverage for the new module, not just a percentage off the list price.

When you demo the secret detection, have them run it against a sample of your own code repos, not just their prepared examples. That's the fastest way to gauge the noise level compared to your current setup.


—HR


   
ReplyQuote
(@harperk)
Honorable Member
Joined: 4 months ago
Posts: 537
 

The dashboard consolidation is the shiny object they want you to chase. The real test is whether the secret detection and IaC scanning share the same rule logic and taxonomy as their cloud alerts, or if you're just getting a prettier UI for three separate tools bolted together.

On your last question, the pricing is always additive, but the discounts can be deceptive. Negotiate the module based on the number of repos or lines of code, not a vague percentage off your existing bill. They'll try to bundle it as a "platform expansion" at a flat fee.

And run that trial against your noisiest, oldest repos. That's how you'll see if the secret detection is actually tuned or just a rebranded open-source scanner.


Data over dogma.


   
ReplyQuote
(@crusty_pipeline_redux)
Honorable Member
Joined: 8 months ago
Posts: 469
 

One fewer dashboard is the classic vendor bait. You'll trade separate tools for a single dashboard that's now got three times the noise.

> How the secret detection compares
Run it against a real repo, not their demo. If it's just a wrapper on a stale Gitleaks fork, you'll know in five minutes.

The IaC scan catching things earlier? That's the only real question. If it doesn't stop the build before the cloud scanner pings you, you're just paying for a prettier alarm bell.


-- old school


   
ReplyQuote
(@benwhite)
Reputable Member
Joined: 4 months ago
Posts: 209
 

The split-brain scenario is inevitable unless you rip out Snyk completely. Their engine will never fully ingest a competitor's data format.

If the policy engine is the real value, you have to go all in. Adding this module while keeping your SCA tool outside means you've just added a third policy engine, not unified anything.


read the fine print


   
ReplyQuote
(@hannahw)
Reputable Member
Joined: 4 months ago
Posts: 234
 

Absolutely agree on the need for a single engine. That's the whole pivot.

But going all-in might not be the only path if your SCA contract has another year on it. We kept ours running in parallel, but suppressed its alerts in favor of the unified ones. It's messy, but worked as a stepping stone without burning the budget.

Have you found a clean way to ingest Snyk findings into another engine, or is that just a myth vendors sell?



   
ReplyQuote
(@davids)
Honorable Member
Joined: 4 months ago
Posts: 568
 

You've got a good handle on the pros and cons, Dan. Your instinct to compare the secret detection directly against something like GitGuardian is spot on. In my experience, the value hinges on whether they're using a fresh ruleset tuned for dev environments, or just repackaging a generic scanner that floods you with old, irrelevant keys.

On your second question about IaC scanning catching things earlier, I'd push for a concrete demo. Ask them to show you the timestamp difference between an IaC scan failure in your pipeline and the same alert from their cloud scanner after deployment. If it's less than the time it takes your average deployment to roll out, the consolidation is worth considering.

For pricing, they'll almost certainly present it as additive. The key is to negotiate based on a new, discrete unit like "per active repository scanned" instead of a vague platform fee. That gives you a clear cost metric for the new coverage, separate from your existing container count.


Stay curious, stay critical.


   
ReplyQuote
(@ethanp23)
Reputable Member
Joined: 4 months ago
Posts: 293
 

Yes on the false positives - that's the real time sink. We ran a trial of a similar module and my team spent a week just triaging old AWS keys from test accounts that their engine flagged as "live." It wasn't a new finding, just noise from a different source.

You're right about proving the IaC scan catches things earlier. We asked for that exact demo, and the "earlier" finding was only about 90 seconds before the cloud alert. For us, that wasn't enough to justify the tuning overhead.


Beta tester at heart


   
ReplyQuote
(@hiroshim)
Noble Member
Joined: 4 months ago
Posts: 767
 

Your 90-second delta on the IaC scan is a critical data point. That aligns with my own benchmarks, where the "earlier" detection often just reflects the pipeline's own execution delay, not a fundamentally faster scan.

The false positive issue with legacy test keys is systemic. Many of these modules fail to correlate the detected secret with the actual environment context, flagging anything that matches a pattern. A useful test is to check if the module can integrate with your cloud provider's metadata to confirm whether a flagged AWS key is even associated with an active IAM user. Most can't.

This turns consolidation from a time-saver into a data reconciliation task. You're not reducing noise, you're shifting its source.



   
ReplyQuote
(@benwhite)
Reputable Member
Joined: 4 months ago
Posts: 209
 

You're right about the pricing. They'll quote the "platform expansion" flat fee to make the math easier. It's a trap. That fee usually hides the real per-seat or per-repo multipliers that hit you on renewal.

Push them to define exactly what's covered in that flat fee. Is it unlimited repos? Unlimited users? Unlimited scans? It never is.


read the fine print


   
ReplyQuote
Page 1 / 2