Wednesday, June 24, 2026 Archive

Archived Security Snapshot

Critical vulnerabilities, curated daily for security professionals

Archived Security Brief

Wednesday's disclosures center on a cluster of remotely exploitable flaws in AI and ML development tooling, including Crawl4AI, Flowise, Langflow, and picklescan, alongside a high-severity issue in Red Hat Ansible Automation Platform 2.5. Critical CVEs rose to 30 (up 76% from 17), while high-priority CVEs fell to 57 (down 30% from 81). Among the most severe are CVE-2026-56274 (CVSS 9.9) in Flowise, CVE-2026-53753 (CVSS 9.8) in Crawl4AI, and CVE-2026-12417 (CVSS 9.8) in Pravel SignUp/SignIn, with remote code execution and unsafe deserialization recurring across the set. Active exploitation is confirmed in five CVEs spanning Splunk Enterprise, Lantronix EDS5000, and Ubiquiti UniFi OS. No patches were available at disclosure time, so teams should prioritize compensating controls and exposure reduction for affected components until vendor fixes ship.

  • AI/ML development tooling is the dominant theme, with critical RCE-class flaws in Flowise (CVE-2026-56274, CVSS 9.9), Crawl4AI (CVE-2026-53753, CVSS 9.8), Langflow (CVE-2026-48519, CVSS 9.6), and picklescan (CVE-2026-56315, CVSS 9.8)
  • Critical CVEs climbed to 30, a 76% increase over the prior day's 17
  • High-priority CVEs dropped to 57, down 30% from 81 the prior day
  • Remote code execution and unsafe deserialization recur across the critical set, including Electron Framework (CVE-2026-54257, CVSS 9.3) and Red Hat Ansible Automation Platform 2.5 (CVE-2026-11807, CVSS 9.6)
  • Patch availability stands at 0% at disclosure, leaving Immich, GeoVision GV-I/O Box 4E, and the AI tooling components without vendor fixes
  • Five CVEs are under active exploitation, affecting Splunk Enterprise and Cloud, Lantronix EDS5000, and Ubiquiti UniFi OS

Immediate action: Prioritize the actively exploited systems first—Splunk Enterprise and Cloud, Lantronix EDS5000, and Ubiquiti UniFi OS—by restricting management access and applying vendor guidance. For the AI/ML tooling and infrastructure flaws (Flowise, Crawl4AI, Langflow, picklescan, Ansible Automation Platform), no patches are available yet, so isolate affected services, limit untrusted input, and monitor for vendor releases.

How to read this brief

CVSS score (e.g. 9.1) — severity from 0–10. Red marks critical (9+), orange high (7–8.9).

Exploitability — how hard the flaw is to attack, read from the CVSS vector:

  • Network / Adjacent / Local / Physical — how close an attacker must get. Network means reachable over the internet.
  • No / Low / High privileges — the access they need first. No privileges means no login required.
  • No interaction / User interaction — whether a victim has to do something (open a file, click a link). No interaction means fully automatable.

The lower the bar on all three, the easier to exploit at scale — “Network · No privileges · No interaction” is the worst case: hit from anywhere, no credentials, no victim action.

Actively exploited — confirmed under attack in the wild (CISA’s Known Exploited Vulnerabilities catalog). Prioritize these regardless of score.

EPSS · Nth percentile — FIRST.org’s estimated chance a flaw is exploited within 30 days. We flag it only in the top 10% — a statistical signal it’s unusually likely to be targeted, separate from whether attacks are confirmed.

💡 Tip: Swipe CVE cards left to ⭐ star, right to ❌ remove

Section Navigation