Threat Intelligence / CVE / CVE-2026-39987

CVE-2026-39987 — Marimo

CISA KEV

CVSS v3.1 9.8 (CRITICAL) · CVSS v4.0 9.3 · EPSS 82.1% (higher than 99.2% of all scored CVEs) · Priority 9/10 · Published 2026-04-08

As of 2026-04-11, CVE-2026-39987 is a CRITICAL-severity vulnerability in Marimo, CVSS v3.1 9.8, EPSS 82.1% (99.2th percentile). It is listed in the CISA Known Exploited Vulnerabilities catalog (added 2026-04-09), with a US federal remediation deadline of 2026-04-11. Threadlinqs Intelligence links 4 tracked threat campaigns to CVE-2026-39987, most recently “NadMesh Botnet Hunts Exposed AI Services (ComfyUI, Ollama, n8n, Open WebUI, Langflow, Gradio) for Credentials and MCP Tool Access”.

Last updated: 2026-04-11

What is CVE-2026-39987?

Critical pre-authentication remote code execution vulnerability in Marimo, an open-source reactive Python notebook platform (~20,000 GitHub stars). The terminal WebSocket endpoint at /terminal/ws in marimo/_server/api/endpoints/terminal.py performs no authentication check, unlike the primary /ws endpoint which invokes WebSocketConnectionValidator.validate_auth(). Any unauthenticated attacker can complete a WebSocket handshake and obtain a full interactive PTY shell executing commands as the Marimo process user — typically root in default Docker deployments. Active exploitation observed just 9 hours 41 minutes after the GitHub security advisory (GHSA-2679-6mx9-h9xc) was published on April 8, 2026. Threat actor from IP 49.207.56.74 conducted scripted PoC validation, manual filesystem reconnaissance, and targeted credential harvesting extracting .env files containing AWS API keys. Endor Labs found 30 of 186 internet-reachable instances (16%) accepted unauthenticated WebSocket handshakes. CISA added to KEV with remediation deadline of April 11, 2026. Fixed in Marimo v0.23.0 (PR #9098, commit c24d4806).

The record classifies CVE-2026-39987 under weakness class CWE-306. Its CVSS v3 base vector states that the flaw is reachable remotely over the network, needs no prior authentication, needs no user interaction, and has high impact on confidentiality, integrity, availability. 1 affected-product entry is recorded, across 1 vendor, listed below. The identifier was first published 158 days ago.

Severity and exploitation probability

CVSS v3.1 base score: 9.8 — CRITICAL
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS v4.0 base score: 9.3
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
EPSS (FIRST): 82.1% probability of exploitation in the next 30 days, higher than 99.2% of all scored CVEs
CISA KEV: Listed since 2026-04-09, federal remediation deadline 2026-04-11
Threadlinqs priority: 9/10 — CISA KEV-listed, which Threadlinqs floors at 9
Published: 2026-04-08, last modified 2026-04-11

Is CVE-2026-39987 being exploited?

CISA added CVE-2026-39987 to the Known Exploited Vulnerabilities catalog on 2026-04-09, which means the agency holds evidence of exploitation in the wild; US federal civilian agencies had to remediate it by 2026-04-11 under BOD 22-01. It currently carries a trending score of 40 in the Threadlinqs vulnerability feed.

Affected products and versions

Marimo: Marimo

How to fix CVE-2026-39987

The record marks a vendor fix as available for CVE-2026-39987. Patch reference: https://github.com/marimo-team/marimo/pull/9098. Vendor advisory: https://github.com/marimo-team/marimo/security/advisories/GHSA-2679-6mx9-h9xc. Because CVE-2026-39987 is KEV-listed, US federal civilian agencies were required to apply the vendor fix, or stop using the product, by 2026-04-11. Apply the vendor fix referenced above to every affected product listed in this record, then confirm the running version against the vendor advisory.

Threat activity tracking CVE-2026-39987

4 tracked threats in the Threadlinqs corpus reference CVE-2026-39987, either in the campaign’s CVE list or as an indicator on the campaign record.

NadMesh Botnet Hunts Exposed AI Services (ComfyUI, Ollama, n8n, Open WebUI, Langflow, Gradio) for Credentials and MCP Tool Access — HIGH · 2026-07-17
Agentic Threat Actor Container Escape — AI Agent-Driven marimo CVE-2026-39987 RCE → Docker Socket → Host Breakout (nsenter) → Kubernetes Secret Store Dump (Sysdig TRT) — CRITICAL · 2026-06-06
First AI-Agent-Driven Cloud Intrusion — Marimo CVE-2026-39987 RCE → AWS Secrets Manager → SSH Bastion → Internal PostgreSQL Exfiltration (Sysdig TRT, 2026-05-10) — HIGH · 2026-05-28
Marimo Pre-Auth RCE via Unauthenticated WebSocket Terminal (CVE-2026-39987) — CRITICAL · 2026-04-12

Sources

Seeded from threat-derived and not yet processed by the Threadlinqs enrichment pipeline, so blank CVSS, EPSS or KEV fields above mean NOT MEASURED rather than measured-absent.

Full detection coverage & IOCs for threats exploiting CVE-2026-39987 are available via the Threadlinqs MCP server (Purple tier). View plans →

← all vulnerabilities · Markdown version · Threadlinqs Intelligence