Softwr

Cybersecurity · head to head

Falco vs Very Good Security

Falco logo

Falco

Cybersecurity

CNCF-graduated runtime threat detection for Linux and Kubernetes using eBPF

From
Free
Rated
-
Very Good Security logo

Very Good Security

Cybersecurity

Tokenisation proxy that keeps card and personal data out of your own systems and out of PCI scope

From
$1000/month
Rated
-

The short version

  • Only Falco has a free tier, so it costs nothing to try first.
  • Each has a real cost: Falco falco detects and alerts but does not block; stopping an attack requires wiring up Falco Talon or your own response tooling, so out of the box a confirmed detection still means a human intervening after the fact.; Very Good Security vGS sits in the live path of every request carrying sensitive data, so its latency and availability become yours, and an outage in the proxy is a payment outage no matter how healthy your own systems are.
  • They diverge on capability: Falco covers eBPF kernel instrumentation, Very Good Security covers Aliasing proxy.
  • Prices and features above were last checked on 31 August 2026.

Where they differ

Only the attributes on which Falco and Very Good Security actually diverge.

Attributes where Falco and Very Good Security differ
AttributeFalcoVery Good Security
Starting priceFree$1000/month
Pricing modelOpen source, no licence feePer month
Free tierYesNo
PlatformsLinux, Kubernetes, Self-hostedWeb, API

Identical on both: user rating (Not yet rated), category (Cybersecurity).

What each one covers

Drawn from each product's published feature list. An absence here means we hold no record of it - not that the product lacks it.

Only in Falco

  • eBPF kernel instrumentation
  • System call rules engine
  • Container and Kubernetes context
  • Default rule set
  • Falcosidekick
  • Falco Talon
  • Plugins framework
  • DaemonSet deployment

Only in Very Good Security

  • Aliasing proxy
  • PCI scope reduction
  • Network tokenisation
  • Processor optionality
  • Card issuing data
  • Vault and access controls
  • Data residency options
  • Compliance artefacts

What people use each for

The jobs each tool is most often brought in to do.

Falco

  • A platform team that needs to know when a shell is opened inside a production container, which image scanning cannot detect because it happens at runtimenot Very Good Security
  • A regulated business required to evidence host and container intrusion detection on Kubernetes nodes for an auditnot Very Good Security
  • A security team wanting a vendor-neutral detection layer whose rules they can read and modify rather than a black-box agentnot Very Good Security
  • A cluster where a compromised dependency might write to sensitive paths or open unexpected outbound connections, and only kernel-level visibility will catch itnot Very Good Security

Very Good Security

  • A marketplace facing its first PCI DSS Level 1 assessment that wants to keep card data off its own estate rather than harden a dozen servicesnot Falco
  • A merchant negotiating with a second acquirer that needs card credentials portable so the negotiation is real rather than theoreticalnot Falco
  • A fintech collecting bank account and identity documents that wants sensitive fields absent from logs, backups and analytics warehouses by constructionnot Falco
  • A card issuer that must display a full PAN in its own mobile app without the app or its backend touching cardholder datanot Falco

Where each one falls short

Documented limitations, not opinions. Every one is a constraint you would hit in normal use.

Falco

  • Falco detects and alerts but does not block; stopping an attack requires wiring up Falco Talon or your own response tooling, so out of the box a confirmed detection still means a human intervening after the fact.
  • The default rule set is noisy in real clusters and generates a large volume of benign matches from normal operational activity; without weeks of tuning, alert fatigue sets in and the team stops reading the feed, which is the usual failure mode.
  • There is no storage, console, search or case management in the project, so a working detection capability means also running Falcosidekick, an event store, a dashboard and alert routing, all of which you build, host and maintain.
  • The modern eBPF driver requires kernel 5.8 or later; older hosts fall back to the legacy probe or the kernel module, which brings driver-building against kernel headers and the operational fragility that comes with it on every kernel upgrade.
  • Per-node syscall instrumentation carries measurable CPU overhead on busy hosts, and the cost scales with syscall volume rather than with cluster size, so the noisiest and most performance-sensitive workloads are exactly the ones that feel it most.

Very Good Security

  • VGS sits in the live path of every request carrying sensitive data, so its latency and availability become yours, and an outage in the proxy is a payment outage no matter how healthy your own systems are.
  • Token portability is the whole selling point yet leaving VGS means migrating tokens back out, a project the vendor has no incentive to streamline, so the lock-in you removed from your acquirer partly moves to VGS.
  • Entry pricing at around one thousand US dollars a month is real money for a pre-revenue fintech, and it buys volume-limited throughput, so cost scales with exactly the growth that made you buy it.
  • Scope reduction is not scope elimination: your QSA still assesses how you integrate, and teams regularly discover that a support tool or an internal admin screen pulled plaintext back in and dragged systems into scope again.
  • Proxy-based interception constrains how you design request flows, and non-standard payloads, streaming uploads or binary formats often need custom routing rules that make debugging production issues noticeably harder.

Pricing, plan by plan

Falco

Free
  • Falco (open source)Free
    • Apache 2.0 licence, CNCF graduated project
    • eBPF and kernel module drivers
    • Full rules engine and default rule set

Very Good Security

$1000/month
  • Starter$1000/month
    • Aliasing proxy
    • Vault storage
    • PCI scope reduction
  • Growth$undefined/month
    • Network tokenisation
    • Multiple processors
    • Data residency options
  • Enterprise$undefined/year
    • Custom vault architecture
    • Dedicated support and SLA
    • Contractual compliance coverage

Which should you pick?

Choose Falco if

  • You need ebpf kernel instrumentation.
  • You want to start without paying.
  • You work on Linux, Kubernetes, Self-hosted.
  • You also want system call rules engine.

Choose Very Good Security if

  • You need aliasing proxy.
  • You work on Web, API.
  • You also want pci scope reduction.

Questions people ask

Is Falco or Very Good Security better?
Neither clearly leads. Falco starts at Free and Very Good Security at $1000/month, and user ratings are close enough to be indistinguishable. Choose on capability and platform support.
Which is cheaper, Falco or Very Good Security?
Falco has a free tier; the other does not. Paid plans start at Free for Falco and $1000/month for Very Good Security.
Does Falco or Very Good Security run on more platforms?
Falco runs on Linux, Kubernetes, Self-hosted. Very Good Security runs on Web, API.
Can I use Falco for free?
Yes. Falco has a free tier, so you can try it without paying. Very Good Security starts at $1000/month.
What is Falco best used for?
Falco is most often used for a platform team that needs to know when a shell is opened inside a production container, which image scanning cannot detect because it happens at runtime, a regulated business required to evidence host and container intrusion detection on kubernetes nodes for an audit, a security team wanting a vendor-neutral detection layer whose rules they can read and modify rather than a black-box agent, a cluster where a compromised dependency might write to sensitive paths or open unexpected outbound connections, and only kernel-level visibility will catch it. Of those, a platform team that needs to know when a shell is opened inside a production container, which image scanning cannot detect because it happens at runtime and a regulated business required to evidence host and container intrusion detection on kubernetes nodes for an audit are not what Very Good Security is typically brought in for.
What can Falco do that Very Good Security cannot?
Falco covers eBPF kernel instrumentation, System call rules engine, Container and Kubernetes context, Default rule set. Very Good Security covers Aliasing proxy, PCI scope reduction, Network tokenisation, Processor optionality.

Answered from the vendors’ own pages

Falco: Does Falco block attacks?

No. It detects and emits events. Response requires Falco Talon or your own automation on top.

Very Good Security: Does VGS make me PCI compliant?

No. It removes cardholder data from your systems so your assessment covers a far smaller boundary, but you still complete an assessment and your integration is part of it.

Falco: Is Falco owned by Sysdig?

Sysdig created and open sourced it, but it graduated within the CNCF in February 2024, so governance sits with the foundation rather than the vendor.

Very Good Security: Can I move to another processor without re-collecting cards?

Yes, that is a core reason people buy it. The vault reveals stored credentials to whichever processor you route to.

Falco: What does it cost?

The project is Apache 2.0 with no licence fee. The cost is the storage, routing, tuning and staff time needed to make its output useful.

Very Good Security: What does it cost?

Published entry pricing is about one thousand US dollars per month; growth and enterprise tiers are quoted.

Falco: What kernel version do I need?

Kernel 5.8 or later for the default modern eBPF driver. Older hosts need the legacy eBPF probe or the kernel module.

Very Good Security: Is it only for card data?

No. The proxy handles any sensitive field, including bank details, national identifiers and documents, though payments is where the product is now focused.

Share

Related pages

Other head to heads