Vulnerability report

Active exploitation observed Confirmed confidence In CISA KEV

CVE-2026-9198

Langflow OSS Unauthenticated Remote Code Execution

IBM / Langflow OSS · 1.0.0 to <= 1.10.0

Severity
CVSS 9.8 · Critical
Confidence
Confirmed
Exploit status
Observed in sensors
EPSS
28.7%
First observed
06 Jul 2026
Last observed
09 Oct 2026

Decision summary

What security teams need to know first

Direct answers before the deeper technical record.

What it is

CVE-2026-9198 is an unauthenticated vulnerability affecting IBM Langflow OSS. IBM Langflow OSS 1.0.0 through 1.10.0 allows unauthenticated attackers to chain /api/v1/auto_login (mints SUPERUSER tokens to any network...

Is it exploited?

Yes. Previdian sensors observed exploitation attempts with confirmed confidence. Listed in CISA KEV. Also confirmed by third-party sources.

Who is affected?

IBM / Langflow OSS 1.0.0 to <= 1.10.0.

What should we do?

Patch immediately, validate internet-facing exposure, and monitor for matching requests.

Overview

Langflow OSS Unauthenticated Remote Code Execution

IBM Langflow OSS 1.0.0 through 1.10.0 allows unauthenticated attackers to chain /api/v1/auto_login (mints SUPERUSER tokens to any network caller) with /api/v1/validate/code (executes user code via exec()) to achieve full RCE on default Langflow deployments

IBM Affected
Langflow OSS
Through 1.10.0
View vendor advisory (opens in new tab)
Published
17 Jul 2026
Exploited Since
06 Jul 2026
Attack vector
Remote
Complexity
Low
Privileges
None
User interaction
None

Tags

cisa nuclei_scanner

CVE References

Exploitation evidence

Why Previdian marks this CVE as exploited

Third-party attestation and first-party sensor observation are shown separately so teams can judge the evidence chain.

Exploited in the wild

CISA

Recorded 04 Aug 2026

Previdian independently recorded this as exploited after first seeing it in honeypot sensors.

Active exploitation observed

Previdian sensor

First observed 06 Jul 2026

Previdian first observed exploitation attempts targeting this vulnerability in our honeypot sensors.

Proof of concept available

GitHub

Recorded 21 Jul 2026

Public scanner or PoC coverage increases practical exploitability.

Known exploited vulnerability sources

Per-source evidence links for KEV attestations are available through the Previdian Pro API.

Learn about Pro API access
Source Added
CISA First 2026-08-04 16:45 UTC
Previdian Sensors 2026-08-04 18:01 UTC
CVE 2026-08-04 18:20 UTC
The Hacker News 2026-08-05 07:40 UTC

Operational indicators for this CVE are listed under Detection.

Sensor telemetry

First-party evidence of exploitation activity

Previdian sensors recorded exploitation attempts targeting this vulnerability. The cards and chart show volume, unique attackers, and daily activity.

1,886

Attempts observed

920

Unique attacker IPs

66

Attacker countries

AE · AR · AT · AU · AZ · BD · BG · BO · BR · BT · CA · CH · CN · CZ · DE · DK · DO · DZ · EC · EE · EG · ER · ES · ET · FI · FR · GB · GH · HK · ID · IE · IN · IQ · IR · IS · IT · JP · KE · KH · KR · LT · MA · MD · MX · MY · NG · NL · PK · PL · PT · RO · RS · RU · SA · SC · SE · SG · TH · TN · TR · TW · US · VN · XK · YE · ZA

52

Sensors observed

CVE-2026-9198 exploitation attempts over the last 91 days

Daily events observed by Previdian sensors

Updated 09 Oct 2026

1
3
11
3
0
2
2
21
6
12
12
23
30
13
51
31
27
41
34
36
53
32
35
34
31
43
32
14
17
42
26
31
23
14
11
7
4
29
27
3
7
19
11
8
15
11
7
14
19
19
10
32
14
23
38
15
12
11
8
8
19
20
41
14
16
23
22
22
27
43
40
34
56
18
11
8
12
15
23
6
13
19
36
17
30
21
25
21
31
31
4
11 Jul 30 Jul 18 Aug 6 Sep 9 Oct

First observed 06 Jul 2026 · Last observed 09 Oct 2026

Pro adds sensor region and window summaries. Enterprise adds raw IPs, paths, User-Agents, and payloads.

Start 14-day Pro trial

Detection

Operational artifacts and observed signals

Make the evidence actionable in scanner, SOC, and edge-control workflows.

Observed signals

Request targets
User-Agents
Callback hosts
9
229
1

Request targets and User-Agents available in Pro. Callback host details available in Enterprise.

Virtual patch status

No Previdian virtual patch is currently available. Future rules ship for ModSecurity, Cloudflare, and AWS WAF.

Learn about virtual patches →

Attacker IP Indicators

Attacker IP indicators observed · available in Pro and Enterprise.

Sensor-derived attacker IP indicators are available to Pro and Enterprise accounts under Detection and through the Pro API.

Learn about Pro API access

Scanner Artifacts

Scanner and exploit-framework references linked to this CVE.

Risk and context

Severity, weaknesses, and research context

CVSS v3.1

9.8 Critical

Potential damage if exploited. Separate from whether attackers are using it.

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

EPSS

28.7%

Estimated chance of exploitation in the next 30 days. Previdian's warning comes from evidence, not this score.

Recent mention · RecordedFuture

August 2026 CVE Landscape

In August 2026, Insikt Group® identified 73 high-impact vulnerabilities that should be prioritized for remediation, 43 of which had a Very Critical Recorded Future Risk Score. This represents a 14% decrease from last month.

Read full advisory

All Mentions

Recent mention · RecordedFuture

August 2026 CVE Landscape

RecordedFuture · 08 Sep 2026

In August 2026, Insikt Group® identified 73 high-impact vulnerabilities that should be prioritized for remediation, 43 of which had a Very Critical Recorded Future Risk Score. This represents a 14% decrease from last month.

Recent mention · The Hacker News

CISA Flags Langflow RCE, Tomcat, and N-central Flaws as Actively Exploited

The Hacker News · 05 Aug 2026

The U.S. Cybersecurity and Infrastructure Security Agency (CISA), on August 5, 2026, added three flaws to its Known Exploited Vulnerabilities (KEV) catalog, citing evidence of active exploitation in the wild. The list of vulnerabilities is as follows - CVE-2026-9198 (CVSS score: 9.8) - A code injection vulnerability in Langflow that allows unauthenticated attackers to achieve full remote

Potential Proof of Concepts

These PoCs are unverified and could contain malware. Use at your own risk.

joaovicdev/EXPLOIT-CVE-2026-9198

github · Created 2026-08-29 23:37:33 UTC · 0 stars · AI assessment 90%

K3ysTr0K3R/CVE-2026-9198

github · Created 2026-08-22 15:55:09 UTC · 0 stars · AI assessment 90%

CVE-2026-9198 - IBM Langflow OSS Unauthenticated Remote Code Execution (RCE)

CuteeCat/CVE-2026-9198

github · Created 2026-08-11 00:55:44 UTC · 0 stars · AI assessment 85%

CVE-2026-9198利用代码

rmhowe425/PoC-CVE-2026-9198

github · Created 2026-08-05 17:44:08 UTC · 1 stars · AI assessment 90%

IBM Langflow - Remote Code Execution

nuclei · Created Unknown

Timeline

From disclosure to observed exploitation

  1. 07:40 UTC

    KEV confirmed by The Hacker News

    Exploitation attested by an external source

  2. 18:20 UTC

    KEV confirmed by CVE

    Exploitation attested by an external source

  3. 16:45 UTC

    Added to CISA KEV

    Listed in the CISA Known Exploited Vulnerabilities catalog

  4. 04:30 UTC

    Public PoC available

    Public proof-of-concept code published

  5. 04:30 UTC

    Nuclei template available

    Scanner coverage available

  6. 17:36 UTC

    CVE published

    Vulnerability disclosed publicly

  7. 14:26 UTC

    Indicators of compromise added (927)

    Indicators of compromise recorded

  8. 14:26 UTC

    Observed by Previdian sensors

    Evidence-backed exploitation signal

  9. 15:40 UTC

    CVE ID reserved

    Identifier reserved by the CNA

Pro API

Automate this intelligence

Confidence, exploit status, sensor telemetry, PoCs, scanner integrations, mentions, and tags are available programmatically for VM, SOC, and CTI workflows.

  • Evidence confidence and provenance
  • First-party sensor telemetry
  • PoC and scanner references
  • Affected versions and enrichment
  • Automation-ready JSON delivery

GET /api/v2/pro/kevs/CVE-2026-9198

Free JSON includes basic KEV fields
{
  "cve_id": "CVE-2026-9198",
  "confidence": "Confirmed",
  "cvss_score": 9.8,
  "cvss_estimated": false,
  "epss_score": 0.28658,
  "exploit_status": {
    "exploited_in_the_wild": true,
    "active_exploitation_observed": true
  },
  "sensor_telemetry": { "attempts": 1886, "sensors": 52 }
}