Products

Continuous risk management platforms CONTINUOUS VISIBILITY

Your perimeter, with automated monitoring, AI triage, analyst validation and tracking of every fix through to verified closure. Your suppliers, measured from the outside and with a list of decisions, each with its reason.

External exposure management

Continuous monitoring of the external perimeter, critical findings reported in under 2 hours and a retest of every fix.

See plans

01 · Scope of the analysis

Six areas of analysis across your external perimeter

All analysis is performed from the outside, read-only, and within the time window your organization authorizes.

/01

Asset inventory

Discovery of internet-facing assets that are not in your inventory: subdomains, ports, services, certificates and the JavaScript code your applications expose.

SubdomainsPortsCertificates
/02

Data exposure

Detection of publicly accessible files, backups and admin panels. Every finding is validated by its content, not just by the server's response.

BackupsPanelsDirectories
/03

Exposed secrets

Identification of keys and credentials published in the code of your sites. Whether they are still valid is checked with the provider through an identity lookup, and they are never used for anything else.

API keysTokensCredentials
/04

Cloud configuration

Review of cloud storage that can be listed or read without credentials, and of the permissions your access policy grants.

BucketsStoragePermissions
/05

Identity and email

Assessment of impersonation risk: SPF, DKIM and DMARC configuration, and detection of subdomains open to takeover.

SPFDKIMDMARC
/06

Exploited vulnerabilities

Correlation of your exposed technology with known vulnerabilities that are under active exploitation, according to the CISA KEV catalog.

CVECISA KEVPrioritization

02 · Methodology

A cycle that ends in verification, not in the report

Most services hand over a list of findings. Our process adds a fourth step: confirming that each fix actually removed the risk.

  1. 01

    Scope definition

    The authorized assets, the time window and the source IP addresses of the analysis are agreed in writing.

  2. 02

    Controlled analysis

    One request at a time per server, read-only, and strictly within the authorized window.

  3. 03

    Expert validation

    Confirmed findings are reported immediately. Doubtful cases are reviewed by an analyst, who confirms or dismisses them with documented reasoning.

  4. What sets us apart04

    Retest and verified closure

    When your team marks a finding as fixed, the platform runs the test again. If the risk is still there, the finding is reopened with the evidence.

AI in the platform

Detection is measured; AI helps decide

Findings come from deterministic checks that can be repeated and explained: no model invents a finding or decides what gets scanned. AI works on what has already been measured, with a person in charge of anything that is not clear-cut.

  • AI triageAn agent reviews every new finding and answers whether it is real, a false positive or uncertain, with its confidence and its reasoning for the analyst. It starts in shadow mode and only publishes or dismisses on its own once its precision, measured against the analyst's decisions, allows it. A critical alert is never dismissed automatically.
  • Analyst copilotIt weighs in on whether a finding is a false positive, suggests a severity and drafts findings that are not in the catalog. It advises the analyst: it never publishes or changes the status of a finding.
  • Agents with human approvalAn agent matches the CISA catalog of known exploited vulnerabilities (KEV) against what your perimeter exposes: when the match is exact, the alert goes out on its own; when it is not, it waits for an analyst. Every agent proposal is logged, and a single brake stops them all.
  • Your data, protected before it leavesAI is enabled only with the client's acceptance in the contract. Before the model is queried, secrets and personal data are masked and your domains and IP addresses are replaced with pseudonyms; the client's name is never sent.

03 · Deliverable

Prioritized findings, with evidence and status

Illustrative sample of a baseline. Every finding includes the affected asset, its severity, the technical description and the remediation status.

baseline · [domain] · sample
AssetSeverityFindingStatus
api-staging.[domain]CriticalAdmin panel reachable from the internetOPEN
[domain]-backupsHighCloud storage allows public listingIN PROGRESS
legacy.[domain]NewLive server missing from the inventoryOPEN
mail.[domain]MediumDMARC policy published but not enforcedREMEDIATED

04 · Plans

Three plans, analyst and retest included

Prices in USD plus applicable taxes. A 12-month contract with monthly billing and 30 days' notice to terminate. Each domain, subdomain or IP in scope counts as one asset; there is no charge per port, per finding or per retest.

Perimeter

USD 590/ month

Implementation and baseline: USD 850, one-time payment

Organizations with one or two domains and no dedicated security team.

  • Up to 2 domains and 60 assets
  • Weekly full active analysis
  • Passive monitoring every 6 hours
  • Analyst and retest included
Sign up for Perimeter
Recommended

Managed

USD 1,340/ month

Implementation and baseline: USD 1,250, one-time payment

Groups with several brands or subsidiaries: mid-size banking, healthcare, retail and education.

  • Up to 6 domains and 200 assets
  • Daily active analysis
  • Scan of all 65,535 ports every quarter
  • Monthly review meeting
Sign up for Managed

Vital

USD 2,850/ month

Implementation and baseline: USD 1,900, one-time payment

Operators of vital importance, holding companies and organizations subject to due diligence.

  • Up to 15 domains and 500 assets
  • Two analysis windows per day
  • Scan of all 65,535 ports every month
  • Documentation for due diligence processes
Sign up for Vital

Implementation and initial baseline, as a single payment: USD 850 on Perimeter, USD 1,250 on Managed and USD 1,900 on Vital. Above 500 assets we prepare a custom quote. Pentest and advisory services are quoted separately.

05 · Frequently asked questions

What security teams usually ask

How is this different from a pentest or an automated scanner?

A pentest gives you a point-in-time view, limited to one or two weeks, and an automated scanner hands over raw results. The platform is continuous: every finding is validated by an analyst or by a technical check, and none is closed without a retest to confirm it.

What information about our organization is processed, and where is it stored?

Only the information your organization already publishes on the internet. Evidence is stored masked: in the case of a credential, we keep its cryptographic fingerprint and never its value.

Is there any risk of disrupting the operation of our systems?

The analysis is limited to read requests, one at a time per server and within the authorized window. No brute-force attacks are run and no exploitation payloads are executed without written authorization, and there is an immediate stop mechanism.

What are the terms for ending the contract?

The contract runs for 12 months, with monthly billing and 30 days' notice to terminate. When it ends, we hand your data back to you and delete it from our systems.

Third-party risk management

Available

External measurement of each supplier from public sources, a security questionnaire the supplier answers and signs, and a list of decisions with their reasons.

In your client panel

What to do about your supply chain, and why

Your organization's security also depends on the security of your suppliers. You propose your suppliers from the panel, ThreadFusion reviews each proposal, and nothing is measured until it is accepted. Once accepted, each supplier is measured from the outside, from public sources only and with no intrusive testing, and what is measured is shown next to what you and the supplier declare, each with its date.

  • Your suppliers, proposed by youEach supplier with its domain and type of relationship: whether it connects to your systems, with its connection points, or provides you with capacity or a service. The form asks for no passwords, tokens or keys.
  • The consequence, declared by youYou declare what happens to your business and to people's wellbeing if a supplier goes down. Nothing comes pre-filled.
  • External measurement per supplierThe email posture of its domain (SPF, DKIM and DMARC), its subdomains visible in public sources and secrets tied to its domain in public GitHub repositories, among other passive checks, each fact with its date. What cannot be measured from the outside is stated, with the reason.
  • Who your suppliers depend onThe third parties your suppliers depend on according to their public DNS records, such as who handles their email or serves their DNS, and which ones several of them share.
  • Supplier questionnaireThe supplier answers and signs a security questionnaire. What it declares is labelled as its own, next to what was measured and with both dates, and it can dispute a measured fact.
  • Decisions and executive reportA prioritized list of decisions about your suppliers, each with its reason and with no scores or traffic lights, and an executive report for the board.

See it in a demo

We show you the panel working and look together at how it would apply to your suppliers. Contracting is agreed directly with ThreadFusion. In the form, choose “Third-party risk management (demo)”.

Request a demo