Vulnerability Disclosure Program: we build it, we run it, we keep it alive

A VDP is the channel through which someone outside your company warns you about a flaw before an attacker uses it. We design and publish it, answer every incoming report within agreed deadlines, and review the programme each quarter so it still covers what you actually run today. No dedicated in-house team needed, and with the documentary evidence NIS2 and the Cyber Resilience Act ask for.

Last updated:

  • 24 h · 72 h · 14 d — CRA deadlines to report an exploited vulnerability: early warning, notification and final report (Art. 14)
  • 5 years — Minimum period the CRA requires manufacturers to handle vulnerabilities and ship security updates
  • ISO 29147 + 30111 — Disclosure and internal handling standards we build the process on
  • 2-4 weeks — From the first meeting to a live policy and a published /.well-known/security.txt

A disclosure channel with nobody behind it is worse than none

Publishing a responsible disclosure policy is the easy part. The hard part starts afterwards, and it almost always breaks in the same three places: the report lands in a mailbox nobody watches and surfaces weeks later; an automated acknowledgment goes out but a real answer never does, so the researcher gives up and publishes the flaw on their own terms; or the report does get read, but nobody can tell whether it is severe, a duplicate, or something a regulator must hear about, so the decision sits waiting.

The cost of those three failures is not only technical. Publishing a disclosure channel is a public commitment, and slow handling reads from the outside as indifference to security. On top of that, now that NIS2 and the Cyber Resilience Act require vulnerability handling and disclosure, what used to be good practice leaves a paper trail: with no record of what came in, when it was answered and what was fixed, there is nothing to show an auditor.

We take on the part that breaks. We design the programme, open the channel, answer every report within agreed deadlines and leave the full record behind.

What the service covers, phase by phase

  • Baseline assessment: reports you already receive, channels already open, and which disclosure obligations actually apply to you
  • Drafting the responsible disclosure policy: scope, rules of engagement, out-of-bounds testing and a safe harbor clause
  • Intake channel implementation: encrypted form, PGP mailbox or specialized platform (HackerOne Response, Bugcrowd, Intigriti)
  • Publication where researchers look for it: /.well-known/security.txt (RFC 9116), security page and public policy
  • Agreed SLAs for acknowledgment, first assessment and resolution, with a dedicated escalation path for critical findings
  • Management of every incoming report: intake, acknowledgment, technical triage, duplicates and closure
  • Researcher communication on your behalf, in English and in Spanish
  • Validated findings prioritized with CVSS 4.0 and EPSS, so your team fixes by real risk instead of arrival order
  • Coordination with your technical team during remediation and verification of the fix before the case is closed
  • Coordinated disclosure management (CVD, ISO/IEC 29147) and CVE requests when the researcher wants to publish
  • Support with regulatory reporting: what falls inside the CRA deadlines and what has to reach the relevant CSIRT
  • Quarterly programme review: scope, SLAs met, policy and volume, plus a monthly activity report

How we implement it

  1. Assessment and policy design: We analyze the environment, agree what is in and out of scope, draft the disclosure policy and set the response SLAs. We follow ISO/IEC 29147 for disclosure and ISO/IEC 30111 for internal handling of the report.
  2. Channel implementation and publication: We set up the intake channel and publish it where people look: security.txt under /.well-known/, your security page and the public policy. We run a test report end to end before announcing it.
  3. Triage operations: Every report follows the same path: acknowledgment to the researcher, technical reproduction, duplicates and false positives ruled out, severity with CVSS 4.0 and EPSS, and handoff to the remediation team with concrete fix steps.
  4. Coordinated disclosure: If the researcher wants to publish, we agree on the timeline, prepare the public communication, handle the CVE request and make sure the patch is available before the details go out.
  5. Programme maintenance: Each quarter we review the scope against what you have shipped since the last review, which SLAs were met and which were not, and the policy if anything relevant changed in regulation or in the product.

Who does what, and how long it takes

The three phases of the service, with the owner of each step and the committed deadline where there is one. Steps without a clock are agreed with your team, not left undated.

We build it

From no channel at all to a published, tested programme in two to four weeks.

  1. Baseline assessment (QuantumSec · Week 1): We review what reports already reach you and through which routes, and which disclosure obligations apply to you under NIS2, the CRA or your sector.
  2. Scope decision (Your team · Week 1): We agree which domains, products and versions are in, what stays out, and which tests are never allowed (denial of service, social engineering against staff, third-party data).
  3. Policy drafting (QuantumSec · Week 2): A full responsible disclosure policy: scope, rules of engagement, what the researcher can expect in return, deadlines and a safe harbor clause reviewed with your legal team.
  4. Intake channel (QuantumSec · Week 3): We build the channel you chose — encrypted form, PGP mailbox or specialized platform — and wire it into our triage flow.
  5. Publication and test (QuantumSec · Week 4): We publish the policy and the security.txt file under /.well-known/ (RFC 9116), and send a test report end to end before the programme is announced.

By the end of this phase: Policy published, channel live, security.txt in place and the whole path verified with a real test report.

We run it

Every report that arrives has an owner, a deadline and an answer. In August too.

  1. The report arrives (Researcher): The researcher submits the finding through the published channel. It lands in our queue, not in a shared mailbox nobody watches.
  2. Acknowledgment (QuantumSec · 24 business hours): A real reply to the researcher confirming receipt, with the case number and the deadline for a first assessment.
  3. Technical triage (QuantumSec · 72 business hours): We reproduce the flaw, rule out duplicates and false positives, and set severity with CVSS 4.0 and exploitation likelihood with EPSS.
  4. Critical escalation (QuantumSec · 4 business hours): If the impact could be critical, we alert your security lead through the agreed emergency route, whatever the time, and coordinate the response.
  5. Remediation (Your team): We hand over the validated finding with fix steps, answer your team’s questions and verify the fix before closing the case.
  6. Closure and coordinated disclosure (QuantumSec): We close the loop with the researcher. If they want to publish, we agree on the timeline, handle the CVE request and prepare the public communication for after the patch.

By the end of this phase: No report left unanswered, anything critical escalated the same day, and a full record of what came in, what was validated and what was fixed.

We keep it alive

A programme nobody reviews goes stale: the scope leaves out whatever you shipped afterwards.

  1. Monthly report (QuantumSec · Monthly): Report volume, validation rate, duplicates, actual response times against the committed ones and the status of every open case.
  2. Scope review (QuantumSec · Quarterly): We go over what you have put into production since the last review and add it to the published scope, so the programme covers what is really exposed.
  3. Policy and SLA tuning (QuantumSec · Quarterly): If deadlines are being missed or regulation has moved, we update the policy, the SLAs and the security.txt file, and republish them.
  4. Audit evidence (QuantumSec): We keep the file a NIS2, CRA or ISO 27001 auditor asks for: current policy, report history, deadlines and the decision taken in each case.
  5. Evolution call (Your team): When the programme takes in more than it can absorb, we assess with you whether to move to a paid bug bounty or to a phased scope.

By the end of this phase: A current programme, with a scope that matches what you have deployed and a compliance file ready for whoever asks.

What you keep

  • A responsible disclosure policy drafted, reviewed and published
  • A safe harbor clause setting out your legal commitment to good-faith researchers
  • A live, secured intake channel with security.txt published per RFC 9116
  • Full management of every incoming report, with the agreed SLAs in writing
  • Every validated finding rated with CVSS 4.0, EPSS and concrete fix steps
  • A complete record of researcher interactions, case by case
  • A monthly activity report and a quarterly programme review
  • A compliance file for NIS2, the Cyber Resilience Act and ISO 27001 audits
  • An agreed, documented coordinated disclosure procedure, ready for the first case

Who this service is for

  • Essential or important entities under NIS2, where vulnerability handling and disclosure is one of the Article 21 measures
  • Manufacturers of products with digital elements under the Cyber Resilience Act, who need a CVD policy and the ability to report within 24 and 72 hours
  • Companies already receiving reports over LinkedIn, the contact form or the support mailbox with no process to handle them
  • Small security teams that cannot guarantee a response out of hours or during holidays
  • Startups that want a serious disclosure channel from day one, without running a paid bug bounty
  • Public bodies and government suppliers that must evidence a vulnerability notification channel
  • Companies with a live bug bounty that need to separate the free VDP from the paid programme

Security mailbox, managed VDP and bug bounty

These three get mixed up constantly. A mailbox is an email address; a VDP is a process with public commitments; a bug bounty adds money to that process to attract more researchers.

security@ mailbox, no processManaged VDPPaid bug bounty
Public policy with scope and rulesNoYes, drafted and publishedYes, plus a bounty table
Legal safe harbor for the researcherNoYesYes
Acknowledgment with a deadlineAutomated or noneA real reply in 24 business hoursDepends on platform or manager
Technical triage of every reportYour team, when it canIncluded, with CVSS 4.0 and EPSSIncluded if triage is managed
Payment to the researcherNoNo: recognition, not moneyYes, per valid finding
Expected report volumeOccasionalLow and steadyHigh, with more noise
Coordinated disclosure and CVE requestsNo processManaged, ISO/IEC 29147Managed
Evidence for NIS2, CRA and ISO 27001Whatever is in the inboxComplete fileComplete file
When it fitsNever as your only measureFirst formal channel and compliance baselineOnce the VDP works and you want more coverage

Compare with

Frequently asked questions about managed VDPs

Does NIS2 require me to have a VDP?

NIS2 never uses the word "VDP", but Article 21(2)(e) includes vulnerability handling and disclosure among the measures essential and important entities must apply, and Article 12 sets up the European coordinated disclosure framework, with a CSIRT designated as coordinator in each Member State and the European vulnerability database maintained by ENISA. For certain entity types — cloud providers, data centres, CDNs, online marketplaces, search engines, managed service and trust service providers, among others — Implementing Regulation (EU) 2024/2690 also spells out the handling and disclosure procedure you must have in writing. In practice, the usual way to meet all of that is a VDP. The assessment tells you which part genuinely applies to you.

What about the Cyber Resilience Act?

The CRA is more explicit: Annex I, Part II requires the manufacturer of a product with digital elements to have a coordinated vulnerability disclosure policy and an accessible channel to receive reports, and to handle vulnerabilities throughout the support period, which cannot be shorter than five years. And since 11 September 2026, Article 14 requires reporting actively exploited vulnerabilities and severe incidents through ENISA's single reporting platform, in three submissions: an early warning within 24 hours, a notification within 72 hours and a final report within 14 days. Without an intake channel and triage able to assess a report within hours, those deadlines are not met.

How long does it take for the VDP to be operational?

Two to four weeks: assessment and scope in week one, policy in week two, channel in week three, publication with an end-to-end test in week four. If an audit or a compliance date is looming it can be compressed, usually by starting with a narrower scope and widening it later.

What is the difference between a VDP and a bug bounty?

A VDP is a responsible disclosure channel: it defines how to report, what is in scope and what the researcher can expect, with no monetary rewards. A bug bounty adds money per valid finding, which attracts far more researchers and, with them, far more volume and far more noise. The logical order is to run the VDP first, check the process holds up, and only then consider a bug bounty. Our bug bounty versus VDP guide compares them in detail.

What happens if we receive a critical vulnerability?

It has its own route. As soon as triage spots that a report could have critical impact, we escalate it straight to your security lead through the agreed emergency channel, whatever the time, and coordinate the response from there. The committed SLA for that alert is 4 business hours, and in parallel we assess whether the case triggers any regulatory reporting obligation.

Do we have to pay rewards in a VDP?

No, and that is exactly what sets it apart from a bug bounty. What the researcher expects in return is something else: that someone answers, that they are told whether the finding was valid, that it gets fixed, and that they are publicly credited in a hall of fame or in the security advisory if they want to be. That side of the deal costs no money; it costs process, and process is exactly what we manage.

What is safe harbor and why does my policy need it?

It is the explicit commitment not to take legal action against anyone who reports a flaw in good faith and within the published scope. Without that clause, a cautious researcher assumes that warning you could cost them dearly and decides not to, or reports anonymously without enough detail to reproduce the issue. We draft the clause and review it with your legal team before it goes out.

Who answers the researchers, you or us?

We answer on your behalf, in English and in Spanish, with your team copied on everything. Decisions that affect the business — accepting a risk, setting a patch date, authorizing a publication — are always yours: we prepare them with the technical context already done so they can be taken quickly.

Can you coordinate the public disclosure of a vulnerability?

Yes. If the researcher wants to publish their finding, on a blog, in a talk or as a CVE, we manage the coordinated disclosure process following ISO/IEC 29147: we agree on the timeline with them, file the CVE request with the relevant CNA, prepare the security advisory and verify the patch is available before the technical details go out.

Start here

What the regulation requires

Running it day to day

Sources

Assess my current situation