The Cyber Resilience Act for Open-Source Maintainers
The EU Cyber Resilience Act is not a blanket rule that makes every volunteer maintainer a product manufacturer. The engineering challenge is to identify the real role in the product chain, then make vulnerability handling and release evidence work across the organizations that build and ship software.
Published September 28, 202615 min readProduct security
Scope note: this is a practical engineering overview, not legal advice. The legal classification depends on the facts: who places a product on the EU market, whether activity is commercial, who has responsibility for the product, and what sustained support an organization provides. Get qualified counsel for a specific product or organization.
Start with role, not repository ownership
The CRA regulates products with digital elements made available on the EU market and places the core product-conformity obligations on manufacturers. Open-source software made available in a commercial activity can fall within scope; software developed or supplied outside a commercial activity is treated differently. A person contributing code to open-source software without responsibility for that software is not automatically the manufacturer.
The Act defines an open-source software steward as a legal person, other than a manufacturer, that provides systematic and sustained support for specific free and open-source software intended for commercial activities, and ensures its viability. That is a narrower, organization-level role, not a synonym for every individual maintainer. A company integrating a library into a product it places on the market may have manufacturer responsibilities for that product even when upstream components are community-developed.
Responsibility follows the role in the product chain, not the Git hosting account
01 / ContributorWrites or reviews codeA contribution alone does not make an individual the product manufacturer. Keep provenance and follow the project’s security process.
02 / StewardOrganization sustains a FOSS productWhere the statutory criteria apply, document a verifiable cybersecurity policy, support vulnerability reporting and handling, and cooperate with authorities.
03 / ManufacturerPlaces a product on the EU marketOwns product-level obligations including risk assessment, technical documentation, conformity, support-period handling and in-scope reporting.
These roles can interact. A steward may coordinate a shared upstream vulnerability process while a downstream manufacturer remains responsible for its product, affected versions, user communications and regulatory reporting. A dependency’s SBOM entry is not a transfer of responsibility, and an upstream patch does not prove that a shipped product has been remediated.
The dates are different for manufacturers and stewards
As of September 28, 2026, the Commission’s implementation materials state that the CRA’s Article 14 reporting obligations for manufacturers apply from September 11, 2026. The corresponding reporting obligations for open-source software stewards start December 11, 2027, when the CRA generally becomes applicable. Do not copy the manufacturer deadline into a maintainer checklist as if every open-source project had the same duty.
Milestone
Who / what
Engineering implication
11 September 2026
Manufacturer reporting under Article 14 applies.
Product teams need a tested path to assess, route and report actively exploited vulnerabilities and severe incidents.
11 December 2027
CRA generally applies; steward reporting duties begin.
Organizations that meet the steward definition should have documented policies and a workable vulnerability process before this date.
11 December 2027
Product-level CRA obligations generally apply.
Manufacturers need product-specific conformity evidence, risk assessment, support and update processes.
For manufacturers, Article 14 sets staged reporting timelines, including an early warning within 24 hours and a fuller notification within 72 hours for in-scope actively exploited vulnerabilities and severe incidents. These are not generic incident-response SLAs for every open-source maintainer. Establish who determines applicability and who submits a report; use the EU Single Reporting Platform when the obligation and process apply.
Turn a disclosure into a traceable response pipeline
One vulnerability record should connect upstream evidence to every affected release
Keep the vulnerability record structured: internal identifier, reporter, receipt timestamp, affected ranges, reproduction, severity rationale, exploit status, coordinating parties, patch commits, release versions, advisory links and closure evidence. Access should be restricted while a fix is coordinated; after publication, retain the rationale and links needed to reconstruct what changed and which products were affected.
Make the policy proportional and operational
Article 24 calls for stewards to document a verifiable cybersecurity policy that promotes secure development and effective handling of vulnerabilities, foster voluntary reporting, address and remediate discovered vulnerabilities, share relevant information and cooperate with market-surveillance authorities. The policy should fit the organization and its arrangements. It is not a requirement to build an enterprise security department around a small project.
Publish a security contactExplain how to report privately, what information helps reproduction, expected acknowledgement and coordinated disclosure preferences.
Define maintainership coverageName primary and backup triage owners, supported branches and an escalation path for periods when volunteers are unavailable.
Keep release evidence togetherLink source revision, review, tests, signed artifacts, provenance and advisory. Do not imply a signature alone establishes security.
Map downstream relationshipsGive integrators a stable advisory feed and affected-version data; distinguish upstream fixes from each manufacturer’s product assessment.
Protect the disclosure windowUse private coordination for unpatched issues, limit access to exploit details and agree a practical publication and release sequence.
Exercise the processRun a tabletop from report receipt to fixed release, downstream notification and evidence capture; record where the handoffs fail.
What I would build
For a maintained library or platform component, I would build a lightweight security operations path around existing project tools: a private intake form or email, an access-controlled issue tracker, an SBOM and dependency index per release, a CI job that records test and provenance evidence, and a signed advisory feed. A small service could correlate package coordinates and version ranges against downstream manifests, then create notification tasks for named product owners.
The design goal is traceability, not a promise that every consumer is known. Measure time to acknowledgement, time to validated affected-version range, time to fixed release, and downstream confirmation coverage. Publish limitations plainly. Keep product classification and statutory reporting decisions with the organization that has the relevant legal role.
Failure modes worth testing
Failure
Signal
Control
Every maintainer is assumed to be a steward
Policy assigns duties to individuals without checking the legal-person criteria.
Record role assumptions and have counsel assess the organization and activity.
Upstream patch is mistaken for product remediation
Advisory closes while shipped products still contain vulnerable versions.
Track package version to product release and require downstream disposition.
Disclosure mailbox has no owner
Reports wait through vacations or land in public issue trackers.
Test primary/backup routing and acknowledgement monitoring.
SBOM exists but cannot be queried
Teams cannot identify affected releases during a short reporting window.
Store machine-readable SBOMs by immutable artifact digest and automate component/version lookup.
Incident clocks start late
Report timestamp is lost between support, security and legal queues.
Persist first-awareness timestamps and escalate potential regulatory events immediately.
In summary
The CRA is a product-security regulation with differentiated roles, not a rule that turns every repository contributor into a manufacturer. Open-source stewards have a tailored set of duties when the statutory definition fits; manufacturers retain product-level obligations. Build the shared mechanics now: private reporting, reproducible triage, dependency and release traceability, coordinated fixes, and clear downstream handoffs. Then validate the exact role and deadlines for each organization with authoritative guidance and legal advice.
Editorial note: This article summarizes EU materials for engineering discussion as of September 28, 2026. It is not legal advice and does not determine whether a particular entity or product falls within the CRA.