Home
/
Platform
/
Cryptographic discovery
Enterprise cryptographic discovery

Find cryptography across your enterprise.

A credible inventory begins with a coordinated discovery strategy. Qinsight brings complementary observation methods together and makes coverage gaps explicit.

Plan a pilot
See how the platform works

A scanner tells you what it can see. A discovery program tells you what remains unseen.

Cryptography appears in active protocols, cloud services, keys, certificates, source code, binaries, file systems, software manifests and third-party evidence. Each source answers a different question. The goal is not maximum scanning, it is authorized, explainable coverage with enough provenance to support a trusted inventory.

Live coverage

Coverage is a number you should be able to see

Illustrative coverage shows which discovery surfaces have been assessed and where authorized collection is still needed.

Known gap in this illustrative scope: the OT segment is not currently covered by an authorized collection method.

Coverage by design

Match the collection method to the surface

A technical matrix buyers can use to evaluate fit and scope a pilot, including where confidence is lower and why.

Surface
What it can reveal
Collection
Typical confidence
01
Network
Protocols, cipher suites, certificates and negotiated behavior
Passive + active
High
02
Cloud + KMS
Keys, certificates, services, policy metadata and usage context
API / credentialed
High
03
Source code
Libraries, crypto calls, configuration and embedded material
Static analysis
Medium
04
Files + hosts
Certificate stores, key material metadata, libraries and configs
Credentialed
Medium
05
CBOM + imports
Declared cryptography from vendors and other tooling
CycloneDX import
Depends on source
06
Security tooling
Existing asset and endpoint context to enrich the record
API
Medium
One discovery run

Make every collection event defensible

A discovery run is more than a scan button. It is an authorized scope, a collection method, a set of observations and a reviewable record of what succeeded, failed or remained out of reach.

Explore a sample investigation
01

Authorize

Define environment, target boundaries, credentials and exclusions.

02

Collect

Run the appropriate connector, scanner or import for each surface.

03

Normalize

Map raw evidence to canonical objects without discarding source detail.

04

Resolve

Deduplicate, flag conflicts and connect observations to systems.

05

Review

Measure coverage, freshness, confidence and unresolved gaps.

Provenance

Keep the "how do you know?" answer attached

Every record carries the collection method, the sensor, the timestamp and the confidence. Evidence that cannot be traced is not evidence.

observation / 8AC2-F41Dconfidence 0.96
Observed condition
TLS 1.2 · RSA key exchange
Review
Observed on
payments-api.prod:443
endpoint
Source
authorized network scan
method
Collection run
RUN-2026-08-06-04
traceable
First / last seen
May 14 / 2 hours ago
history
Linked service · owner
Payments gateway · Payments platform
context
Payload retained
no
data boundary

Illustrative record. Example data.

No agents required

Network and cloud collection needs no software on target hosts. Filesystem scanning uses a stateless container with a read-only service account.

Least-privilege by default

Integrations request describe and list permissions. Never decrypt, never export.

Metadata only

Endpoints, certificate fields, protocol details and findings. Never private keys, never payloads.

A focused first proof

Start with one business service and map it completely

A pilot that discovers ten thousand objects across an unclear scope proves less than one that fully maps a single business service. Three things make the difference.

Representative scope
One business service, bounded and agreed before collection starts.
Three to five sources
Enough method diversity to test correlation, not just collection. One source proves a connector works; several prove the record holds together.
Explicit success criteria
Agreed by your team before collection starts. If nobody wrote down what good looks like, the result is always arguable.