Join our Newsletter — 33% off our NHI Course
Home› Guides› EU Cyber Resilience Act Vulnerability and Incident Reporting…
Guide Regulations & Frameworks

EU Cyber Resilience Act Vulnerability and Incident Reporting Guide: Article 14 Deadlines and the ENISA Platform

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 27 October 2026 12 min read
On this page

Since 11 September 2026, every manufacturer selling a product with digital elements in the EU must report actively exploited vulnerabilities and severe incidents within 24 hours, even though the rest of the Cyber Resilience Act (CRA) does not apply until December 2027. Article 14 of Regulation (EU) 2024/2847 is the first CRA obligation to bite, and it covers products already on the market, including legacy versions still in use. Reports go once, through ENISA's Single Reporting Platform (SRP), to the national CSIRT acting as coordinator. This guide goes deeper than the EU Cyber Resilience Act Guide: it covers what triggers a report, the content of each stage, how the platform works in practice, penalties, and the product security process (PSIRT) a manufacturer needs to meet the deadlines. It is written for product security, PSIRT, legal and engineering leads at hardware and software manufacturers, importers and open-source projects.

Status line (checked 27 October 2026): Article 14 applies from 11 September 2026 and the SRP went live that day at portal.cra-srp.enisa.europa.eu. The CRA as a whole applies from 11 December 2027. Commission Delegated Regulation (EU) 2026/881, adopted on 11 December 2025, sets the conditions for delaying dissemination of notifications. The SRP has no API and accepts mandatory notifications only; voluntary reporting under Article 15 is planned for a later phase.

Key takeaways

  • Two triggers, three stages. Actively exploited vulnerabilities and severe incidents each need an early warning within 24 hours, a notification within 72 hours and a final report (14 days after a fix for vulnerabilities, one month after the notification for incidents).
  • Legacy products are in scope. Article 69(3) applies the reporting duty from 11 September 2026 to all in-scope products, including those placed on the market before December 2027, but not to exploitation the manufacturer already knew about before that date.
  • One report, one platform. Manufacturers submit once through the SRP to the CSIRT designated as coordinator in the Member State of their main establishment; ENISA receives it at the same time and the CSIRT shares it with other affected Member States.
  • The highest fine tier applies. Breaches of Article 14 can attract fines up to EUR 15 million or 2.5% of worldwide annual turnover, whichever is higher, though micro and small enterprises are not fined for missing the 24-hour early warning.
  • Access is personal and manual. Submitters are Assigned Representatives with personal EU Login accounts and MFA, up to 21 per manufacturer, entering reports by hand: plan rosters, cover and backups now.

Scope: who and what is covered

Article 14 applies to manufacturers of products with digital elements made available on the EU market: hardware and software products and their remote data processing solutions, within the CRA's scope. Open-source software stewards have a narrower duty under Article 24(3) from 11 December 2027, to the extent they are involved in developing such products. Importers and distributors do not file Article 14 reports, but they must tell the manufacturer when they learn of a vulnerability and are part of the CSIRT selection chain for non-EU manufacturers.

TriggerMeaningNot reportable
Actively exploited vulnerability (Art. 14(1))A vulnerability for which there is reliable evidence that a malicious actor has exploited it in a system without the owner's permissionVulnerabilities found and tested in good faith for research, correction or disclosure (recital 68); unexploited vulnerabilities, which follow Article 13 and Annex I vulnerability handling
Severe incident (Art. 14(3) and (5))An incident that negatively affects, or could affect, the product's ability to protect the availability, authenticity, integrity or confidentiality of sensitive or important data or functions, or that has led or could lead to malicious code being introduced or executed in the product or a user's systemIncidents with no effect on product security, such as a marketing site outage

Recital 68 gives the textbook severe incident: malicious code inserted into the channel used to release security updates. The SolarWinds supply chain compromise, where tampered Orion builds were signed and shipped as legitimate updates, is exactly that pattern.

Which CSIRT receives the report

  1. The CSIRT designated as coordinator in the Member State of the main EU establishment, meaning where cybersecurity decisions are mainly taken (Art. 14(7)); if unclear, the establishment with the most employees.
  2. No EU establishment: the Member State where the authorised representative acts for the most products, then where the importer places the most products, then where the distributor makes the most products available, then where the product has the most users.

ENISA warns that a notification sent to the wrong coordinator may be invalidated and must be resubmitted, which can cost the 24-hour window.

Key obligations

StageActively exploited vulnerabilitySevere incident
Early warning: 24 hours from awarenessWhere applicable, the Member States where the product has been made availableWhether the incident is suspected to be caused by unlawful or malicious acts, and the Member States concerned
Notification: 72 hours from awarenessGeneral product information, the nature of the exploit and vulnerability, corrective or mitigating measures taken and those users can take, and how sensitive the manufacturer considers the informationNature of the incident, initial assessment, measures taken and available to users, and sensitivity
Intermediate reportOn request of the CSIRT (Art. 14(6))On request of the CSIRT
Final reportNo later than 14 days after a corrective or mitigating measure is available: description including severity and impact, information on the malicious actor, details of the security updateWithin one month of the notification: detailed description, severity and impact, likely threat type or root cause, applied and ongoing mitigation
  • Informing users (Art. 14(8)): after becoming aware, tell impacted users, and where appropriate all users, about the vulnerability or incident and the mitigations they can apply, where appropriate in a structured, machine-readable format. There is no fixed deadline, but if users are not informed in a timely manner the coordinating CSIRT may do it. CSAF advisories are a natural fit; see the CSAF Security Advisories Guide.
  • Sensitivity and delay: the manufacturer can mark particular exceptional circumstances in the 72-hour notification. Dissemination to other CSIRTs is then delayed under Article 16(2) and Delegated Regulation (EU) 2026/881, and ENISA initially receives only partial information.
  • SRP data fields: ENISA's glossary lists fields such as CVE and EUVD identifiers, exploitation details, severity and impact, and the malicious actor. Have them mapped before an event; see the CVE, CWE and CPE Guide.
  • Relationship with Article 13: Article 14 sits on top of the CRA's vulnerability handling requirements, which apply in full from December 2027. Reporting without a working intake, triage and patch process is not sustainable.

When does the clock start?

The CRA does not define awareness in detail. Commentary on the Commission's draft guidance describes it as the point where the manufacturer has a reasonable degree of certainty that a vulnerability is being actively exploited or that a severe incident has compromised product security. Unverified rumours do not start the clock, but manufacturers cannot wait for forensic certainty. Record when information arrived, what was known, when the threshold was met and who decided.

Timeline and current status

DateEvent
23 October 2024Regulation (EU) 2024/2847 adopted
20 November 2024Published in the Official Journal
10 December 2024Entry into force
11 December 2025Commission adopts the delegated act on conditions for delaying dissemination (published as Delegated Regulation (EU) 2026/881)
11 September 2026Article 14 reporting applies; SRP operational
11 December 2027CRA applies in full, including essential requirements and Article 24(3) for open-source stewards

The SRP is version one. ENISA notes that the 72-hour counter in the portal currently runs from the early warning submission rather than from awareness, so a report can show as overdue early; the legal deadline is what counts. If the SRP is unavailable, manufacturers may contact their CSIRT directly but must still submit through the SRP once it is restored.

Enforcement and penalties

  • Fines (Art. 64): non-compliance with the essential requirements and with the obligations in Articles 13 and 14 sits in the top tier: up to EUR 15 million or 2.5% of total worldwide annual turnover for the preceding financial year, whichever is higher. Member States set the penalty rules and the market surveillance authorities apply them.
  • Carve-outs (Art. 64(10)): microenterprises and small enterprises are not fined for missing the 24-hour early warning deadline, and open-source software stewards are not fined for any infringement. Operational, contractual and reputational consequences remain.
  • CSIRT and market surveillance follow-up: national CSIRTs may share information with market surveillance authorities, which can require corrective action or withdrawal of non-compliant products.
  • Parallel regimes: one event can also trigger GDPR, NIS2, DORA and sector rules. The SRP does not replace those channels. See the Cyber Incident Reporting Requirements Map and the NIS2 Directive Guide.

Check your national implementing law for when fines can be imposed for Article 14 breaches, since penalty rules are set nationally.

What it means for AI agents and non-human identities

Many severe incidents under Article 14(5) start with a machine identity: a stolen code signing key, a CI/CD token or a cloud credential behind the product's update service. A compromised identity in the build or release path can introduce malicious code into the product, which is the second limb of the severe incident test.

  • Signing and release credentials: theft of a code signing key or release token is a candidate severe incident even before abuse is proven, because it "could lead" to malicious code. See the Compromised Code Signing Certificate Playbook.
  • Pipeline identities: scoped, short-lived CI/CD credentials reduce the chance that a token leak becomes a reportable incident. The tj-actions/changed-files compromise showed how a stolen bot token let attackers push malicious code to an Action used by more than 23,000 repositories. See the CI/CD Pipeline Identity Security Guide.
  • Product-embedded credentials: hard-coded keys, default device certificates and backend API keys shipped in products become actively exploited vulnerabilities the moment someone abuses them in the field.
  • AI agents in the product: agentic features with tool access widen the impact of an exploited flaw. Prompt injection that leads an agent to execute code in a user's environment may meet the "malicious code executed" test.
  • SRP accounts are identities too: Assigned Representatives use personal EU Login with MFA. Treat them as privileged accounts: named, backed up, reviewed on staff change.

Compliance steps

  1. Inventory products: list every in-scope product and version on the EU market, including legacy and unsupported versions still in use, with the responsible legal entity for OEM and white-label lines.
  2. Fix the reporting route: confirm the main establishment and coordinating CSIRT, register a Primary Assigned Representative and Secondary representatives, and test log-in out of hours.
  3. Define triggers: write criteria for "reliable evidence of exploitation" and the Article 14(5) severity tests, with examples, and name who decides.
  4. Wire in intelligence: feed CISA KEV, EUVD, CSIRT advisories, customer reports and bug bounty submissions into PSIRT triage so exploitation evidence is seen within hours. See the Bug Bounty and Vulnerability Disclosure Guide.
  5. Template each stage: pre-build early warning, notification and final report templates mapped to SRP fields, plus user advisories in CSAF.
  6. Align regimes: one incident commander, one fact base, separate submissions for CRA, NIS2, GDPR and DORA.
  7. Exercise: run a tabletop where a field exploit is reported on a Saturday morning and the patch is ready on day ten.
  8. Roll back safely: keep the ability to pull or revoke an update and to rotate signing keys, so a severe incident can be contained as well as reported.

Framework mapping and measurement

CRA reporting dutyNIST CSF 2.0NIST SP 800-53 Rev 5
Detect exploitation of product flawsID.RA-01RA-5, SI-5
24/72-hour reports to CSIRTRS.CO-02, RS.CO-03IR-6
Corrective measures and final reportPR.PS-02SI-2
Incident plan with third partiesRS.MA-01IR-8
Release integrity and provenanceGV.SC-07SR-3, SR-4, SA-11
  • Time to awareness decision: hours from first signal to a documented reportability decision.
  • Early warning timeliness: percentage of reportable events with an early warning inside 24 hours; target 100%.
  • Fix lead time: days from awareness to corrective measure, which drives the 14-day final report.
  • User advisory latency: hours from corrective measure to published advisory.
  • Representative coverage: number of trained Assigned Representatives reachable in each time zone.

Common failure patterns

  • Ignoring legacy versions: assuming the CRA starts in 2027 and missing that reporting already covers products on the market.
  • Waiting for certainty: treating the 24-hour early warning as a confirmed finding instead of a timely alert.
  • Wrong CSIRT: filing with the CSIRT of a sales subsidiary rather than the main establishment, forcing resubmission.
  • Single-person dependency: one registered representative who is on leave when the exploit lands.
  • Counting the portal timer: relying on the SRP counter instead of the legal deadline from awareness.
  • Missing the identity angle: not treating a stolen signing key or pipeline token as a potential severe incident.

Practitioner checklist

  • Confirm which products and versions are in scope today, including legacy lines.
  • Document the main establishment and coordinating CSIRT.
  • Register a Primary and several Secondary Assigned Representatives with MFA.
  • Write reportability criteria for exploited vulnerabilities and severe incidents.
  • Map SRP fields into early warning, notification and final report templates.
  • Connect KEV, EUVD and researcher intake to PSIRT triage with a 24-hour SLA.
  • Prepare machine-readable user advisories.
  • Treat signing keys and pipeline tokens as in-scope incident sources.
  • Exercise the process twice a year and after major product changes.

Standards and references

Related NHI Mgmt Group resources: EU Cyber Resilience Act Guide · Vulnerability Disclosure Maturity Model · SBOM Guide · Code Signing Security Guide. This guide summarises the reporting obligations of the EU Cyber Resilience Act as at 27 October 2026 and is not legal advice.

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 27 October 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.