MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. Starting in 3.3.0 and prior to 3.15.0, the unauthenticated POST /api/2.0/mlflow/webhooks/{id}/test endpoint calls _validate_webhook_url() in mlflow/utils/validation.py only for the original URL while mlflow/webhooks/delivery.py follows redirects and re-resolves the hostname without pinning the validated address, allowing attackers to reach internal or cloud metadata services and receive response_status and response_body. This issue is fixed in version 3.15.0.
CVE-2026-64849
“—” beside the score: it is equal to the CVSS score beside it.
Score elevated to 9.3 because this CVE is listed on the CISA Known Exploited Vulnerabilities catalog (added 2026-08-19), indicating real-world exploitation has been confirmed by US federal agencies. A secondary CVSS source baseline 9.3 retained for reference (NVD's own analysis pending). Confidence: HIGH.
Also known asEUVD-2026-60473 (ENISA EUVD) · GHSA-7gwp-5pfp-969j (GitHub Security Advisories)Source: ENISA EUVD
- Actively exploited in the wild (CISA-KEV) — the reason this is sooner than the SSVC decision on this page at low or medium mission impact
A fix is available — apply it.
- CVSS v3
- 9.3
- EG Score
- 9.3CRITICALhigh confidence
- EG Risk
- 97EG Risk 97/100
EG Risk is EchelonGraph's 0–100 priority score. It fuses impact severity (EG Severity, the CVSS base score, never the KEV-floored or EPSS-lifted EG Score) with real-world exploitation and automatability, under weights EchelonGraph set by hand, so you can rank equal-severity CVEs and fix the most dangerous first. Higher = act sooner.
How it’s computed (value × EchelonGraph’s weight)Severity93% × 45%Impact severity (EG Severity): CVSS 9.3 of 10. Never the KEV-floored or EPSS-lifted EG Score.Exploitation100% × 40%Fixed at 100% for a CISA KEV listing: EchelonGraph's rule.Automatability100% × 15%CISA's SSVC Automatable value: yes (Vulnrichment).SSVC (CISA's decision table, applied by EchelonGraph): Attend at low or medium mission impact; Act at high (mission-essential systems).Exploitation active (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE.Action: A fix is available. Apply it sooner than your standard update timelines at low or medium mission impact and as soon as possible at high. - EPSS PROB
- 10%
- EPSS %ILE
- 95th
- KEV
- ⚠ Exploited
What the EG score is made of (scorer v4.0: the three questions the 0–10 score mixes, published apart · methodology)
The EG score (9.3): severity, escalated by confirmed exploitation. This CVE's 9.3 is its impact severity (EG Severity), not escalated.
- EG Severity
- 9.3 CRITICALCVSS-B, impact only: no KEV listing, EPSS or EG-KEV tier raises it.GitHub Security Advisory CVSS
- EG Likelihood
- 10% within 30 daysSource: FIRST EPSS (model v2026.06.15, score date 2026-11-01). FIRST's exploitation probability, not an EchelonGraph model.
- Confirmed exploitation
- CISA KEV and EG-KEV Tier 1 (exploited in the wild)Reported apart from the likelihood: an observed fact, not a probability.
- EG Priority
- SSVC Attend at medium mission impactAttend at low or medium mission impact; Act at high (mission-essential systems).CISA BOD 26-04: 3 days if the affected system is publicly exposed; 14 days if it is not.Each input and its source, and whom BOD 26-04 binds: the SSVC and BOD 26-04 blocks below.
SSVC (CISA's decision table, applied by EchelonGraph)Attend at low or medium mission impact; Act at high (mission-essential systems).
A fix is available. Apply it sooner than your standard update timelines at low or medium mission impact and as soon as possible at high.
Exploitation active (CISA Vulnrichment) · Automatable yes (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). All three are CISA's SSVC values (Vulnrichment). EchelonGraph applied CISA's decision table to them; CISA publishes SSVC inputs and the table, not a decision for each CVE. Mission impact is CISA's Mission & Well-being decision point, and only you can judge it. CISA's table makes it high when Mission Prevalence is Essential — the vulnerable component "directly provides capabilities that constitute at least one MEF for at least one entity" (MEF: mission essential function) — or when Public Well-Being Impact is Irreversible: "multiple fatalities are likely", the cyber-physical system "is likely lost or destroyed", "extreme or serious externalities" are imposed on other parties, or social systems such as elections or the financial grid "are destabilized and potentially collapse". CISA's decision table
BOD 26-04 (CISA's remediation timeline, Table 1)3 days if the affected system is publicly exposed; 14 days if it is not.
The score's inputs: In CISA KEV yes (CISA's KEV catalog) · Automatable yes (CISA Vulnrichment) · Technical impact partial (CISA Vulnrichment). Whether your system is publicly exposed is yours to answer. BOD 26-04 binds US Federal Civilian Executive Branch agencies; for anyone else it is CISA's published timeline, a reference rather than an obligation. The directive · how EchelonGraph applies it
CISA required action“Apply mitigations in accordance with vendor instructions, ensuring compliance with CISA’s BOD 26-04 Prioritizing Security Updates Based on Risk (see URL in Notes) guidance and CISA’s “Forensics Triage Requirements” (see URL in Notes). Follow applicable BOD 26-04 guidance for cloud services or discontinue use of the product if mitigations are unavailable. Stakeholders are responsible for evaluating each asset's internet exposure and ensuring adherence to BOD 26-04 patching guidelines.”
Quoted verbatim from CISA's Known Exploited Vulnerabilities catalog. CISA directs it at US federal civilian executive branch agencies under Binding Operational Directive (BOD) 26-04, which replaced BOD 22-01 on 10 June 2026; it is not EchelonGraph's advice. CISA's notes: https://github.com/mlflow/mlflow/pull/24258 · https://github.com/mlflow/mlflow/issues/24179 · https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk · https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk · https://nvd.nist.gov/vuln/detail/CVE-2026-64849
Published
August 17, 2026
Last Modified
October 26, 2026
Advisory Details (6)
Auto-updated Oct 26, 2026Known Exploited Vulnerabilities Catalog | CISA
Known Exploited Vulnerabilities Catalog | CISA. Listed in CISA Known Exploited Vulnerabilities catalog.
https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-64849Unauthenticated full-read SSRF in MLflow webhook delivery: _validate_webhook_url bypassed via unvalidated HTTP redirects (and DNS rebinding) · Advisory · mlflow/mlflow · GitHub
https://github.com/mlflow/mlflow/security/advisories/GHSA-7gwp-5pfp-969jv3.15.0
Patch available: mlflow/mlflow v3.15.0
https://github.com/mlflow/mlflow/releases/tag/v3.15.0Fix DNS-rebinding SSRF bypass in webhook delivery
Fix merged in mlflow/mlflow PR #24258 on 2026-07-02 — awaiting tagged release
https://github.com/mlflow/mlflow/pull/24258DNS rebinding bypass of webhook SSRF protection enables cloud metadata access · Issue #24179 · mlflow/mlflow · GitHub
https://github.com/mlflow/mlflow/issues/24179commit ba949522477c (mlflow/mlflow)
Fix landed in mlflow/mlflow commit ba949522477c — awaiting tagged release
https://github.com/mlflow/mlflow/commit/ba949522477cbd5915aa55d29b0cfad7d5ddf939Vendor Advisories for CVE-2026-64849(3)
These vendors published their own advisory mentioning this CVE — often with vendor-specific remediation steps + affected product lists not in NVD.
- RHSA-2026:60520Red Hat Product SecurityCritical
Red Hat Security Advisory: RHOAI 3.4.4 - Red Hat OpenShift AI
- RHSA-2026:60367Red Hat Product SecurityCritical
Red Hat Security Advisory: RHOAI 3.5 - Red Hat OpenShift AI
- GHSA-7gwp-5pfp-969jGitHub Security AdvisoriesCritical
MLflow: Unauthenticated full-read SSRF in webhook delivery: _validate_webhook_url bypassed via unvalidated HTTP redirects (and DNS rebinding)
Patch Availability(3)
| Vendor / Ecosystem | Fix (by version range) / Patch | Released | Source |
|---|---|---|---|
| redhat | rhoai/odh-mlflow-rhel9:1787226790 | 2026-08-27 | redhat |
| redhat | rhoai/odh-latency-predictor-training-rhel9:1786552250 | 2026-08-26 | redhat |
| pip | mlflow @ 3.15.0 (fixes >= 3.3.0, < 3.15.0) | — | ghsa |
Patches are aggregated from vendor advisories (Red Hat, Microsoft, Cisco, GitHub) and package ecosystems (OSV, GHSA). Multiple rows for the same upstream release have been deduplicated.
Affected Packages
(1 across 1 ecosystem)
PyPI(1)
| Package | Vulnerable range | Fix by version range | Dependents |
|---|---|---|---|
| mlflow | 0.0.1 ... 3.9.0rc0 (182 versions) |
| — |
Weakness Classification(1)
MITRE Common Weakness Enumeration — the root-cause categories this CVE belongs to.
Data Freshness Timeline
(refreshed 22× in last 7d / 133× in last 30d)
Each row is a time one of our pipelines fetched, updated or re-scored this CVE. For example, "NVD update" means NVD published or revised its analysis for this CVE; "MITRE cvelistV5" means we ingested or refreshed it from the CNA feed. An arrow marks a change the record states ("CVSS v3 → 9.1"); EPSS and GHSA rows show the values recorded at that time, not what changed (EPSS values are rounded; hover one for the exact figures). Most recent first.
Showing the most recent 100 of 451 refreshes on record for this CVE.
A repeat, a row identical to one the same pipeline wrote less than 10 seconds earlier, is not a separate refresh and is not counted. None are among the latest 100 rows. The 351 oldest rows on record are not shown and were not checked, so the counts can still include repeats.
- 2026-11-01 19:05 UTCEG score recompute
- 2026-11-01 19:05 UTCVendor advisory
- 2026-11-01 19:05 UTCGHSA enrichmentGHSA-7gwp-5pfp-969j · GitHub severity CRITICAL · GitHub CVSS 9.3
- 2026-11-01 16:35 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-11-01 07:40 UTCGHSA enrichmentGHSA-7gwp-5pfp-969j · GitHub severity CRITICAL · GitHub CVSS 9.3
- 2026-11-01 07:40 UTCEG score recompute
- 2026-10-31 18:08 UTCEG score recompute
- 2026-10-31 18:08 UTCVendor advisory
- 2026-10-31 16:33 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-30 17:11 UTCEG score recompute
- 2026-10-30 17:11 UTCVendor advisory
- 2026-10-30 16:35 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-29 16:33 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-29 16:15 UTCEG score recompute
- 2026-10-29 16:15 UTCVendor advisory
- 2026-10-28 16:33 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-28 15:18 UTCEG score recompute
- 2026-10-28 15:18 UTCVendor advisory
- 2026-10-27 16:32 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-27 14:22 UTCEG score recompute
- 2026-10-27 14:22 UTCVendor advisory
- 2026-10-26 13:27 UTCVendor advisory
- 2026-10-25 18:30 UTCEG score recompute
- 2026-10-25 18:30 UTCVendor advisory
- 2026-10-25 16:33 UTCEPSS rescoreEPSS 10% · 95th percentile
Show 75 moreShow fewer
- 2026-10-25 14:12 UTCEG score recompute
- 2026-10-24 17:34 UTCEG score recompute
- 2026-10-24 17:34 UTCVendor advisory
- 2026-10-24 16:33 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-23 16:38 UTCEG score recompute
- 2026-10-23 16:38 UTCVendor advisory
- 2026-10-23 16:32 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-22 16:35 UTCEPSS rescoreEPSS 10% · 95th percentile
- 2026-10-22 15:40 UTCEG score recompute
- 2026-10-22 15:40 UTCVendor advisory
- 2026-10-21 16:33 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-21 14:43 UTCEG score recompute
- 2026-10-21 14:43 UTCVendor advisory
- 2026-10-20 16:33 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-20 13:47 UTCEG score recompute
- 2026-10-20 13:47 UTCVendor advisory
- 2026-10-19 16:32 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-19 12:51 UTCEG score recompute
- 2026-10-19 12:51 UTCVendor advisory
- 2026-10-18 16:32 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-18 11:54 UTCVendor advisory
- 2026-10-18 03:22 UTCEG score recompute
- 2026-10-17 16:35 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-17 10:57 UTCEG score recompute
- 2026-10-17 10:57 UTCVendor advisory
- 2026-10-16 16:43 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-16 09:59 UTCEG score recompute
- 2026-10-16 09:59 UTCVendor advisory
- 2026-10-15 16:32 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-15 09:02 UTCEG score recompute
- 2026-10-15 09:02 UTCVendor advisory
- 2026-10-14 16:33 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-14 08:05 UTCEG score recompute
- 2026-10-14 08:05 UTCVendor advisory
- 2026-10-13 16:34 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-13 07:08 UTCEG score recompute
- 2026-10-13 07:08 UTCVendor advisory
- 2026-10-12 16:34 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-12 15:47 UTCEG score recompute
- 2026-10-12 06:11 UTCVendor advisory
- 2026-10-12 01:48 UTCVendor advisory
- 2026-10-11 21:29 UTCVendor advisory
- 2026-10-11 19:32 UTCEG score recompute
- 2026-10-11 17:08 UTCVendor advisory
- 2026-10-11 16:47 UTCEG score recompute
- 2026-10-11 14:08 UTCEG score recompute
- 2026-10-11 13:49 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-11 12:47 UTCEG score recompute
- 2026-10-11 12:47 UTCVendor advisory
- 2026-10-11 08:27 UTCVendor advisory
- 2026-10-11 04:06 UTCVendor advisory
- 2026-10-10 23:46 UTCEG score recompute
- 2026-10-10 23:46 UTCVendor advisory
- 2026-10-10 20:00 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-10 19:26 UTCVendor advisory
- 2026-10-10 15:06 UTCVendor advisory
- 2026-10-10 10:46 UTCVendor advisory
- 2026-10-10 06:25 UTCVendor advisory
- 2026-10-10 02:05 UTCVendor advisory
- 2026-10-09 21:44 UTCEG score recompute
- 2026-10-09 21:44 UTCVendor advisory
- 2026-10-09 20:07 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-09 17:24 UTCVendor advisory
- 2026-10-09 13:05 UTCVendor advisory
- 2026-10-09 08:43 UTCVendor advisory
- 2026-10-09 04:20 UTCVendor advisory
- 2026-10-09 00:00 UTCVendor advisory
- 2026-10-08 19:41 UTCEG score recompute
- 2026-10-08 19:41 UTCVendor advisory
- 2026-10-08 17:13 UTCEPSS rescoreEPSS 9.8% · 95th percentile
- 2026-10-08 15:21 UTCVendor advisory
- 2026-10-08 11:01 UTCVendor advisory
- 2026-10-08 06:41 UTCVendor advisory
- 2026-10-08 02:21 UTCEG score recompute
- 2026-10-08 02:21 UTCVendor advisory
Public detection templates
(1 reference)A public Nuclei detection template is referenced for this CVE. A detection template checks whether a system is affected; it is not necessarily exploit code.
- Nucleihttp/cves/2026/CVE-2026-64849.yamlPublished Aug 18, 2026
MLflow Webhook SSRF - Unauthenticated Full-Read via Redirect Bypass
Open source ↗
Related CVEs(same product + same vendor + same CWE)
Same product
10 shownPyPI:mlflow
- CVE-2025-15379EG 10.0CRITICAL
- CVE-2025-15036EG 10.0CRITICAL
- CVE-2024-0520NVD 8.8EG 10.0CRITICAL
- CVE-2023-6831NVD 8.1EG 10.0CRITICAL
- CVE-2023-6018NVD 9.8EG 10.0EPSS p99CRITICAL
- CVE-2023-6015NVD 7.5EG 10.0EPSS p91CRITICAL
- CVE-2023-3765EG 10.0EPSS p99CRITICAL
- CVE-2023-2356NVD 7.5EG 10.0EPSS p91CRITICAL
- CVE-2026-116692EG 9.8CRITICAL
- CVE-2026-0545EG 9.8EPSS p91CRITICAL
Same vendor
10 shownredhat:RHSA-2026:60367 · redhat:RHSA-2026:60520
Frequently asked(6)
What is CVE-2026-64849?
When was the CVE record for CVE-2026-64849 published?
Is CVE-2026-64849 actively exploited?
What is the CVSS score of CVE-2026-64849?
Which products are affected by CVE-2026-64849?
How do I remediate CVE-2026-64849?
Dependency Blast Radius
See which npm, PyPI, Go, and Maven packages are affected by CVE-2026-64849
Is Your Infrastructure Affected by CVE-2026-64849?
EchelonGraph automatically scans your cloud infrastructure and maps CVE exposure using blast radius analysis.