Interfaz en español. Las fichas, los resúmenes y las fuentes conservan su idioma original.

Identificá qué requiere atención.

Alertas prioritarias y resúmenes de seguridad. Evidencia, sistemas afectados y próximos pasos.

Editorial methodology

Sources and methodology

Radar connects incident reporting to decisions about your own systems. It does not establish whether your organization was compromised.

How a source becomes an update

  1. 01

    Check original sources

    Vendor advisories, government notices and original research. Secondary coverage can identify a lead; it does not replace the source for fixed versions.

  2. 02

    Compare the evidence

    Review affected products, exploitation, fixes and changed recommendations. Automated checks flag possible changes for review.

  3. 03

    Review before publishing

    Drafts may use AI assistance. A human reviewer must approve technical claims, priorities and actions before publication.

  4. 04

    Keep a visible record

    Corrections stay on the original page. The editorial record explains what changed. Checking a source does not silently rewrite the article.

How response priority works

Act now identifies urgent review or action, such as reported exploitation, compromised credentials or a deadline. An exception to the priority rules is explained in the editorial record.

Review this week identifies planned review of important fixes. Monitor covers developments that need more evidence. A missing patch does not by itself make an issue low priority.

What determines the order

The advisory catalog opens with Latest update: the most recent editorial changes first. This is not the date an attack began, and checking a source alone does not change the editorial update date. Use Response priority to review the action queue described below. Filters and sorting do not change the assigned priority.

When sorted by Response priority, within each priority Radar lists confirmed exploitation first, followed by credential exposure, observed attempts, time-sensitive service impact, working research exploits and vendor precautions. Critical fixes and regional research follow in their respective priority groups. Reviewed CISA KEV matches count as confirmed exploitation within the assigned priority. Equal cases use the editorial deadline, latest update and permanent incident ID; CISA federal due dates do not automatically set a customer deadline.

This is an editorial queue, not a calculated probability of compromise. Check the exposure requirement and affected versions before assigning work. A deadline does not establish exploitation, and a CVSS score does not show whether your systems are affected.

CISA Known Exploited Vulnerabilities

Radar matches CVE identifiers against a reviewed snapshot of CISA’s KEV catalog. The CISA KEV label means that at least one CVE in the incident is listed; the incident page identifies the exact matches and the catalog check date. Inclusion establishes known exploitation, not compromise of your organization. Absence from the snapshot does not rule out exploitation.

The catalog addition date is not the date of the first attack. “Not confirmed” for ransomware means CISA lacks confirmation, not that ransomware use has been excluded. Federal due dates and forensic triage requirements are provided as context; they are not universal obligations for private organizations or businesses in Latin America.

Automated checks identify new entries and changes for editorial review. They do not create public incidents, alter editorial priorities or advance source verification dates. Public evidence remains at the displayed snapshot until a reviewed update is published. Vendor advisories remain the source for affected versions and fixes.

What our labels mean

“Attacks confirmed” means the reviewed source establishes exploitation or the CVE is listed in CISA KEV. “Attack attempts observed” does not establish success. “No attacks reported” applies only to the sources reviewed; “Attacks not established” means the evidence is insufficient, including when a report does not identify which CVE was targeted. These states remain distinct.

In the evidence explorer, response priority appears once for each parent advisory. Linked products and CVEs do not receive independent priorities. Each CVE retains its own evidence and vendor guidance; a finding for one CVE does not apply to the others. Counts show advisories with CVEs, a subset of the full catalog.

Exploit demonstration and observed exploitation are recorded separately. A demonstrated exploit establishes that an attack worked in the documented configuration; it does not establish attacks in the wild. An incident can show both a demonstrated exploit and observed attempts. “Public demonstration not established” means the reviewed sources do not establish one, not that no exploit exists. Each demonstration record identifies its CVEs, source and scope; it does not apply automatically to every CVE grouped in an incident. Vendor confirmation applies to the reported incident, not to the security of your environment.

Record reviewed dates the complete editorial review. Latest evidence check dates the newest individual source or exploit-evidence check; it does not imply that every claim was rechecked. Automatic signal refreshes have separate dates. If a source cannot be reached, that is a gap to resolve, not evidence that the incident is over.

Prioritize vulnerabilities

This view includes current vulnerability advisories. Known exploitation comes first: an exact CISA KEV match or exploitation confirmed for a single CVE by the advisory sources. A claim about a group of CVEs is not assigned to every CVE. Research demonstrations and observed attempts remain distinct from confirmed exploitation.

Among the remaining CVEs, current EPSS scores sort from highest to lowest. Missing, unavailable or stale forecasts appear last and never become zero. EPSS must have a successful check within 48 hours and a score dated within two UTC calendar days. Confirmed exploitation retains precedence even with a low or missing EPSS score. Ties use the parent advisory response order and the CVE identifier.

Rows refer to individual CVEs; response actions and editorial priority belong to the linked advisory. A vulnerability without an assigned CVE remains visible with no EPSS score. Search for products you use, then check the affected versions and configuration. This view does not establish your exposure or scan your environment. The general catalog keeps Latest update as its default.

EPSS, regional advisories and package records

FIRST EPSS is a dated forecast of observed exploitation over the next 30 days. It is not a customer risk score, and a low probability does not override confirmed exploitation. Percentiles rank scores against other CVEs. Missing scores are shown as unavailable, not zero.

Regional coverage uses official notices from CSIRT Panamá, CERT-PY and CISC Brasil. Countries identify the issuing authority, not confirmed victim locations. We link notices to incidents only when CVE identifiers match exactly. Original titles retain their source language. Indexes and RSS feeds have limited coverage and can omit publication dates.

OSV queries use explicitly mapped package names, ecosystems and versions. OpenSSF source records are checked when OSV attributes a record to that repository. An imported OSV record and its OpenSSF original are one piece of evidence, not independent corroboration. Withdrawn records remain labeled, and a no-match response does not certify package safety.

Automated source checks support editorial review. If a check fails, Radar retains the last available data and its original date. Source updates require review before publication; they do not automatically change response priorities or trigger email alerts.

FIRST EPSS · OSV · OpenSSF Malicious Packages · Structured source data

Public data fields

Schema version 1 remains compatible. cve_details assigns vendor, products, source and optional CVSS to each CVE. sector identifies a target industry; it is not a software vendor. services lists available assessment services. latest_evidence_checked_at dates the latest individual check; last_verified_at retains the complete record review date.

Exploitation unknown means evidence is insufficient; none_known means the reviewed sources report no exploitation. Exploit demonstrations remain separate. Patch status describes the available response and does not certify remediation in your environment. Regional counts show grouped notices with publication dates in the last 30 days of the snapshot; older and undated notices remain in the archive and relevant incident records.

RSS publication dates remain stable when records change. JSON Feed supplies separate publication and modification dates. Existing EPSS history fields are retained for compatibility; signals.json is the complete source snapshot.

Corrections and scope

Send corrections to info@oktsec.com with the item URL and supporting source. Do not send secrets or private incident data. Radar is a selected editorial collection, not continuous monitoring of your environment.