CriticalCVE-2026-9198

    Unauthenticated Remote Code Execution via Auto-Login Bypass and Code Validation (CVE-2026-9198)

    Published Last modified In CISA KEV since Sources: CVE.org, pwn.watch, SingCERT, BSI

    What is affected, what to do

    The vulnerability CVE-2026-9198 allows unauthenticated attackers to achieve full remote code execution (RCE) on default Langflow deployments through a code injection vulnerability. This is accomplished by chaining the /api/v1/auto_login endpoint, which generates SUPERUSER tokens, with /api/v1/validate/code, which executes user code. It affects Langflow versions 1.0.0 to 1.10.0 and is actively exploited in attack campaigns. Given its listing in the CISA KEV catalog, an immediate update to version 1.10.1 is recommended to mitigate the risk.

    The vulnerability is being actively exploited; treat it as a priority. Update langflow to version 1.10.1, Langflow Desktop OSS to version 1.10.1. Immediate actions, mitigation and detection rules are in the runbook below.

    Affected
    • IBM Langflow OSS lte 1.0.0..1.10.0

    Not sure whether and how you are affected? Ask a question about this vulnerability

    Metrics

    9.8

    Priority

    XIEM Effective CVSS: highest available CVSS rating from NVD and vendor data.

    Source: NVD, CVSS v3

    9.8

    CVSS v3

    Severity Critical on the scale from 0 to 10.

    Source: CVE.org, pwn.watch, SingCERT, BSI

    28.7 %

    EPSS score

    Elevated — model estimates 10-50% exploitation likelihood.

    Source: FIRST EPSS, percentile 98.1

    Show all metrics
    Severity
    Critical
    Priority (0–325)

    The priority score is a consolidated metric that condenses several individual signals into a single, action-guiding ranking: technical severity (CVSS), exploitation likelihood (EPSS), known-exploited status (KEV) and the context of the affected asset. The aim is to answer the only truly important question in vulnerability management: what first?

    Read in glossary
    247.09 Methodology
    Exploit maturity

    An exploit is code or a technique that specifically uses a concrete vulnerability to force unintended behaviour — such as executing code or escalating privileges. It is the tool that turns a theoretical flaw into a practical attack.

    Read in glossary
    actively exploitedSource: VulnCheck KEV (opens in a new tab), CSA/SingCERT How we classify
    XIEM Effective CVSS

    XIEM® is the tiered managed-security operating model from NEOSEC and a registered trademark of NEOSEC GmbH. It extends classic SIEM — the central collection, normalisation and correlation of security-relevant events — with endpoint forensics, network visibility and deception technology, and consolidates them into three building tiers: Sentry, Orchestrate and Command.

    Read in glossary
    9.8 Methodology
    CVSS v3

    CVSS is the globally established, open standard for assessing the technical severity of a vulnerability on a scale from 0 to 10. It describes how easily a flaw can be exploited and what damage it can cause — but not how likely its actual exploitation is. CVSS and the complementary EPSS deliberately answer different questions.

    Read in glossary
    9.8
    EPSS score

    The EPSS score is the absolute model output for an individual vulnerability: the estimated probability of real-world exploitation within the next 30 days, expressed as a value between 0 and 1 (i.e. 0 % to 100 %). A score of 0.42 means „the model estimates a 42 % chance of real-world exploitation within 30 days“.

    Read in glossary
    28.7 %
    Elevated — model estimates 10-50% exploitation likelihood.
    EPSS percentile

    The EPSS percentile is the relative position of a vulnerability within all EPSS-scored CVEs, expressed as a value between 0 and 1. A percentile of 0.90 means „this CVE has a higher EPSS score than 90 % of all other scored CVEs“. Unlike the EPSS score, it is not a probability — but a positioning within a ranking.

    Read in glossary
    98.1
    Critical — this CVE ranks in the top fifth of all CVEs scored today (rank ≥ 80%).
    KEV status

    The KEV catalogue is a list maintained by the US authority CISA of vulnerabilities for which active exploitation in the wild has been confirmed. It answers not the question „how severe?“ but „is it already being attacked?“ — thereby sharply prioritising patching.

    Read in glossary
    Actively exploited
    Published
    In CISA KEV since

    The KEV catalogue is a list maintained by the US authority CISA of vulnerabilities for which active exploitation in the wild has been confirmed. It answers not the question „how severe?“ but „is it already being attacked?“ — thereby sharply prioritising patching.

    Read in glossary
    CWE

    CWE is a catalogue of typical weakness types in software and hardware — such as flawed input validation, unsafe memory handling or poor authentication. Where a CVE names a concrete vulnerability in a concrete product, a CWE describes the underlying class of defect.

    Read in glossary
    CWE-94

    Response and mitigation

    Why act now?

    Prioritisation rationale

    This flaw is network-reachable without authentication or user interaction. Chaining the auto-login endpoint with the code-validation endpoint yields full code execution on default deployments, which makes it ideal for automated mass scanning. Exploitation likelihood is very high, placing it among the CVEs most likely to be exploited. CISA does not currently flag known ransomware use. Typical victims are quickly stood-up AI/LLM prototypes in data-science and innovation teams that often sit outside normal asset and patch management while concentrating valuable API keys and data access. Rate any instance reachable from the internet or from broad internal networks as critical and act immediately. Locally bound Desktop installs rank lower but still need patching.

    1. Runbook step 1

      Immediate response (0-24 h)

      • Pull exposed instances off the internet now: Inventory every Langflow deployment (OSS, Docker, Desktop). Firewall off anything publicly reachable, whether on the default port 7860 or behind a reverse proxy. The exploit chain needs no authentication, so treat every reachable instance as potentially compromised.
      • Patch: Upgrade Langflow OSS or Langflow Desktop to 1.10.1 or later (pip install -U "langflow>=1.10.1" or the current container image). Versions 1.0.0 through 1.10.0 are affected.
      • Kill auto-login: Set LANGFLOW_AUTO_LOGIN=False regardless of patch level. Define a dedicated superuser with a strong password via LANGFLOW_SUPERUSER and LANGFLOW_SUPERUSER_PASSWORD, then restart the service.
      • Rotate secrets: Assume everything stored in Langflow is burned. This includes LLM provider API keys (OpenAI, Azure, watsonx, etc.), vector-DB and database credentials, LANGFLOW_SECRET_KEY, and any host/pod environment variables or cloud tokens.
      • Preserve evidence, then rebuild: For instances that were reachable before patching, snapshot the container or VM and keep the Langflow database and proxy logs. Rebuild from a clean image rather than cleaning in place.
      • Check reporting duties: If compromise is confirmed, start the NIS2 clock (early warning to the national CSIRT/BSI within 24 h).

    2. Runbook step 2

      Mitigation layers

      • Network segmentation: Keep Langflow on an internal segment or VLAN. Allow access only through VPN/ZTNA with an authenticating front end (e.g. OAuth2-Proxy backed by your IdP). Apply an egress allowlist so the host can reach only the LLM endpoints it needs, which blocks reverse shells and payload staging.
      • WAF/reverse-proxy rule: Until patched, block /api/v1/auto_login and /api/v1/validate/code at the proxy. For nginx: location ~ ^/api/v1/(auto_login|validate/code) { return 403; }. Add a WAF signature that drops POST bodies to these paths containing import os, subprocess, __import__, exec( or eval(.
      • Host/container hardening: Run Langflow as an unprivileged user with a read-only root filesystem, no Docker socket mount and no --privileged. Apply cap_drop: ALL and enable seccomp/AppArmor profiles.
      • IAM/least privilege: Strip the host's or service account's cloud permissions to the minimum. Do not auto-mount Kubernetes service-account tokens. Store only narrowly scoped, spend-capped API keys inside Langflow.
      • Configuration: Bind Langflow to 127.0.0.1 instead of 0.0.0.0 when a local reverse proxy fronts it. Lock down CORS.

    3. Runbook step 3

      Detection rules

      • Web server/proxy logs: A hit on /api/v1/auto_login from an unexpected source, followed within minutes by POST /api/v1/validate/code from the same IP. Example SPL: index=proxy uri_path IN ("/api/v1/auto_login","/api/v1/validate/code") | stats values(uri_path) dc(uri_path) as n by src_ip | where n=2
      • EDR process ancestry: The Langflow Python/uvicorn process spawns sh, bash, curl, wget, nc, python -c or base64. Sigma shape: ParentImage|endswith: python* AND ParentCommandLine|contains: langflow AND Image|endswith: [/sh, /bash, /curl, /wget, /nc]
      • Linux auditd/Sysmon for Linux (EID 1/3): execve events whose ppid is the Langflow process, plus new outbound connections from it. Also watch for writes to ~/.ssh/authorized_keys, /etc/cron* or /tmp.
      • Network telemetry (Zeek/Suricata): HTTP requests where uri is /api/v1/validate/code and the body contains Python imports. Follow up on subsequent outbound connections to unusual ports or to cloud metadata endpoints (169.254.169.254).
      • Application layer: Audit the Langflow database for unknown superuser accounts, freshly minted API keys, or flows with embedded custom code created outside change windows.

    Detection rules

    No NEOSEC detection rule is available for this vulnerability yet.

    Classification under NIS2 and CRA

    Rule-based classification from exploitation status and severity. It does not replace a case-by-case assessment.

    NIS2

    NIS2 is EU Directive (EU) 2022/2555 to raise the level of cybersecurity across the Union. It substantially broadens obligations compared with the first NIS directive: more sectors, mandatory risk management, staged reporting duties and personal accountability of senior management.

    Read in glossary
    Actively exploited: treat the vulnerability as a priority in vulnerability handling under Art. 21(2)(e) NIS2. If it is exploited in your environment, check the reporting obligation under Art. 23 (early warning within 24 hours).
    CRA

    The CRA is EU Regulation (EU) 2024/2847 setting horizontal cybersecurity requirements for „products with digital elements“ — from industrial controllers and software to connected consumer devices. For the first time manufacturers must ensure security across the entire product lifecycle, handle vulnerabilities and provide security updates.

    Read in glossary
    Since 11.09.2026, manufacturers of affected products with digital elements must report actively exploited vulnerabilities under Art. 14 CRA, with an early warning within 24 hours via ENISA's single reporting platform.
    ISO/IEC 27001

    ISO/IEC 27001 is the internationally leading standard for an information security management system (ISMS). It describes a risk-based, process-oriented framework to systematically plan, implement, review and improve information security.

    Read in glossary
    Control A.8.8, management of technical vulnerabilities: assess relevance to your systems, plan remediation and keep evidence.

    Description

    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

    Source: NVD (NIST), pwn.watch, cvelistv5, CISA KEV Catalog, CVE.org (MITRE), CSA/SingCERT, BSI WID Portal, BSI CSAF

    Affected products

    The following versions are affected. Older releases of the same product line are also vulnerable unless stated otherwise.

    Affected products
    Affected versionsFixed in
    IBMLangflow OSS
    • 1.0.0 – ≤ 1.10.0
    • —
    langflowlangflow
    • 1.0.0 – < 1.10.1
    • 1.10.1
    IBMLangflow Desktop OSS
    • < 1.10.1
    • 1.10.1

    Vendors and products

    Weakness classes (CWE)

    • CWE-94Improper Control of Generation of Code ('Code Injection')

      Base

      The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment.

      Entry at MITRE CWE-94

    Public exploit references

    Publicly reported proof-of-concepts, exploit modules and detection templates for this vulnerability. NEOSEC links to the original sources only and does not provide any exploit code.

    Public exploit references
    CISA KEV CatalogCISA Known Exploited Vulnerability (opens in a new tab)cisa.govactively exploited
    CSA/SingCERTActive exploitation of Critical Vulnerability in IBM Langflow OSS (opens in a new tab)CSA/SingCERTactively exploited
    PoC-in-GitHub (nomi-sec)langflow-cve-2026-9198 (opens in a new tab)github.comPoC (publicly reported)reported, not verified
    PoC-in-GitHub (nomi-sec)PoC-CVE-2026-9198 (opens in a new tab)github.comPoC (publicly reported)reported, not verified
    PoC-in-GitHub (nomi-sec)CVE-2026-9198 (opens in a new tab)github.comPoC (publicly reported)reported, not verified
    PoC-in-GitHub (nomi-sec)CVE-2026-9198 (opens in a new tab)github.comPoC (publicly reported)reported, not verified
    PoC-in-GitHub (nomi-sec)CVE-2026-9198 (opens in a new tab)github.comPoC (publicly reported)reported, not verified
    PoC-in-GitHub (nomi-sec)CVE-2026-9198 (opens in a new tab)github.comPoC (publicly reported)reported, not verified
    PoC-in-GitHub (nomi-sec)CVE-2026-9198 (opens in a new tab)github.comPoC (publicly reported)reported, not verified

    References & sources

    • ibm.comIBM Langflow — CVE-2026-9198Vendor advisory, Patch, pwn-watch, Vendor advisory
    • cisa.govknown-exploited-vulnerabilities-catalog?field_cve=CVE-2026-9198government-resource, US government resource
    • pwn.watchpwn.watch: IBM Langflow: IBM Langflow Code Injection Vulnerabilitypwn-watch
    • NVD (NIST)NVD: CVE-2026-9198pwn-watch
    • cisa.govCISA Known Exploited Vulnerabilities Catalogpwn-watch

    Changes to this CVE

    Who changed this CVE and when: NVD (analysis, affected products, CVSS), the CVE publisher (CNA) and additional providers such as CISA (ADP).

    1. Record modifiedIBM

      1 individual change: Affected versions (publisher)

      Show the original change
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Affected versions (publisher)changed
      Langflow OSS
      Langflow OSS

      Source identifier: psirt@us.ibm.com

    2. NVD analysis modifiedNVD

      Show the original change
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Reference typeadded
      —
      CISA-ADP: https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-9198 Types: US Government Resource

      Source identifier: nvd@nist.gov

    3. Record modifiedCISA ADP

      CISA assessment (SSVC): exploitation: active, automatable: yes

      Show 2 original changes
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Referenceadded
      —
      https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-9198
      SSVC assessmentchanged
      {"id":"CVE-2026-9198","role":"CISA Coordinator","options":[{"exploitation":"none"},{"automatable":"yes"},{"technicalImpact":"total"}],"version":"2.0.3","timestamp":"2026-07-17T00:00:00+00:00"}
      {"id":"CVE-2026-9198","role":"CISA Coordinator","options":[{"exploitation":"active"},{"automatable":"yes"},{"technicalImpact":"total"}],"version":"2.0.3","timestamp":"2026-08-04T17:42:22.802585Z"}

      Source identifier: 134c704f-9b21-4f2e-91b3-4a467353bcc0

    4. Record modifiedCISA ADP

      CISA assessment (SSVC): exploitation: none known, automatable: yes

      Show the original change
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      SSVC assessmentchanged
      {"id":"CVE-2026-9198","role":"CISA Coordinator","options":[{"exploitation":"none"},{"automatable":"yes"},{"technicalImpact":"total"}],"version":"2.0.3","timestamp":"2026-07-17T17:52:19.835444Z"}
      {"id":"CVE-2026-9198","role":"CISA Coordinator","options":[{"exploitation":"none"},{"automatable":"yes"},{"technicalImpact":"total"}],"version":"2.0.3","timestamp":"2026-07-17T00:00:00+00:00"}

      Source identifier: 134c704f-9b21-4f2e-91b3-4a467353bcc0

    5. ReceivedIBM

      CVSS V3.1: AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

      Weakness type: CWE-94

      Description added or changed

      Show 5 original changes
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Affected versions (publisher)added
      —
      Langflow OSS
      Descriptionadded
      —
      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
      CVSSadded
      —
      AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
      Weakness type (CWE)added
      —
      CWE-94
      Referenceadded
      —
      https://www.ibm.com/support/pages/node/7278927

      Source identifier: psirt@us.ibm.com

    Show one older change

    1. Record modifiedCISA ADP

      CISA assessment (SSVC): exploitation: none known, automatable: yes

      Show the original change
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      SSVC assessmentadded
      —
      {"id":"CVE-2026-9198","role":"CISA Coordinator","options":[{"exploitation":"none"},{"automatable":"yes"},{"technicalImpact":"total"}],"version":"2.0.3","timestamp":"2026-07-17T17:52:19.835444Z"}

      Source identifier: 134c704f-9b21-4f2e-91b3-4a467353bcc0

    One NVD translation into another language is hidden.

    Linked advisories

    Embed badge

    Show the severity and KEV status of this vulnerability on your site – with a link back here. The image is updated hourly.

    Badge preview for CVE-2026-9198
    <a href="https://intel.neosec.eu/en/cve/CVE-2026-9198"><img src="https://intel.neosec.eu/badge/cve/CVE-2026-9198.svg?lang=en" alt="CVE-2026-9198 – NEOSEC Intel" height="20"></a>