critical severityConfidence highHistorical record

Langflow OSS 33-CVE disclosure: no-login auth bypass and code execution through 1.12.8 (CVE-2026-93680, CVE-2026-110186), fixed in 1.13.0

Direct answer

Langflow OSS 33-CVE disclosure: no-login auth bypass and code execution through 1.12.8 (CVE-2026-93680, CVE-2026-110186), fixed in 1.13.0, also tracked as CVE-2026-93680 / CVE-2026-110186 / CVE-2026-104339 is a critical-severity threat campaign HOL Guard tracked from Oct 22, 2026 to Oct 29, 2026, verified against 13 primary sources and last reviewed Oct 29, 2026. Active tracking has ended and the record remains published as a permanent historical research record.

On 2026-10-29 between 15:14 and 15:19 UTC, IBM (the CNA for Langflow) published 33 CVE records for Langflow OSS, backed by three IBM security bulletins dated 2026-10-19 (nodes 7293015 and 7293016) and 2026-10-28 (node 7294467). Four IDs affect every release from 1.0.0 through 1.12.8 and are fixed only in 1.13.0 (released 2026-10-22): CVE-2026-93680 (9.8, improper authentication, no login), CVE-2026-110186 (9.8, code execution via improper control of code generation, no login), CVE-2026-110193 (8.8, a low-privileged user makes the Agentic Assistant generate or submit component code that runs OS commands) and CVE-2026-110192 (6.5, insufficiently protected credentials). Sixteen IDs through 1.12.3 are fixed in 1.12.4, including the unauthenticated CVE-2026-104339 (9.8) and message-broker deserialization CVE-2026-104317 (8.1). Thirteen through 1.12.4 are fixed in 1.12.5, including three authenticated 9.9 code-execution bugs (CVE-2026-107327, 107328, 107329). Release builds of 1.12.x default LANGFLOW_AUTO_LOGIN to true, which signs visitors in as the bootstrap superuser, so account-required bugs are reachable by anyone who can open the UI on a default install. Langflow 1.13.0 turns auto-login off by default and requires a superuser password. A Metasploit auxiliary module for CVE-2026-93680 (released 2026-10-23) opens the MCP SSE endpoint at /api/v1/mcp/sse without credentials; the module lists 1.5.0 through 1.11.4.

Also tracked as CVE-2026-93680 · CVE-2026-110186 · CVE-2026-104339 · CVE-2026-107327 · CVE-2026-107328 · CVE-2026-107329 · CVE-2026-110193 · Langflow 1.13.0 · Langflow OSS is affected by multiple vulnerabilities · Langflow auth bypass · Langflow unauthenticated RCE · Langflow MCP SSE auth bypass · IBM Langflow security bulletin

First observed
Oct 22, 2026
Last observed
Oct 29, 2026
Last reviewed
Oct 29, 2026
Tracking ended
Oct 29, 2026

Record exports

JSON recordArtifact CSVSTIX 2.1 bundle

Uncertainty and limitations

Read the evidence limits before acting on this record.

IBM published one-line descriptions with no endpoint names or proof of concept, and its 1.13.0 bulletin summary does not map its themes (incomplete auto-login remediation, MCP tool enumeration RCE, API key exposure, Agentic Assistant code execution) to specific IDs. The Metasploit module lists 1.5.0 through 1.11.4 for CVE-2026-93680 while IBM lists 1.0.0 through 1.12.8. No in-the-wild exploitation of any of the 33 had been reported and none was in CISA KEV (catalog 2026.10.28) as of 2026-10-29 15:50 UTC. CVSS scores are IBM's.

  • Servers already running Langflow 1.13.0 are not affected by any of the 33 IDs
  • Instances bound to 127.0.0.1 are not reachable by the no-login network bugs, though the Agentic Assistant issue still applies to anyone using it
  • Instances with auto-login off and only trusted accounts remove most of the account-required risk but not CVE-2026-93680, CVE-2026-110186 or CVE-2026-104339
  • Deployments behind an authenticating proxy covering every Langflow path require the attacker to pass that layer first
  • Releases before 1.0.0 are not listed by IBM, which does not mean they are safe; they carry older KEV-listed bugs

Guard coverage and policy

What Guard's reviewed assertions do and do not claim for this campaign, and the reviewed starting points for defense.

Not covered, or partially covered

  • not covered

    HOL Guard reviews AI coding agent actions on developer machines. It does not patch or inspect Langflow servers; the fix is upgrading to Langflow 1.13.0 and restricting network access.

No campaign-specific policy guidance is published.

Safe next step. Campaign guidance is defensive context. It does not auto-apply policy or claim universal campaign coverage.

Open policy controlsCampaign-specific emergency policy starting point

Timeline

Reviewed events, oldest first, each tied to a verified source.

  1. Vendor action

    Langflow 1.12.4 ships to PyPI; IBM later lists it as the fix for 16 of the 33 IDs.

    Sources:langflow 1.13.0 on PyPI·IBM Security Bulletin 7293015: Langflow OSS 1.0.0-1.12.3, fixed in 1.12.4

  2. Vendor action

    Langflow 1.12.5 ships; IBM lists it as the fix for 13 more IDs including three 9.9 authenticated RCEs.

    Sources:langflow 1.13.0 on PyPI·IBM Security Bulletin 7293016: Langflow OSS 1.0.0-1.12.4, fixed in 1.12.5

  3. Vendor action

    Langflow 1.13.0 ships with automatic login disabled by default; it is the only release fixing the four IDs that run through 1.12.8.

    Sources:langflow-ai/langflow v1.13.0 release notes·langflow 1.13.0 on PyPI·IBM Security Bulletin 7294467: Langflow OSS 1.0.0-1.12.8, fixed in 1.13.0

  4. Update

    Rapid7's Metasploit wrap-up announces an auxiliary module for CVE-2026-93680 that initializes MCP sessions over /api/v1/mcp/sse without credentials.

    Sources:Rapid7: Metasploit Wrap Up: Langflow, MongoDB, and More·Metasploit module langflow_auth_bypass_cve_2026_93680

  5. Disclosure

    IBM publishes 33 Langflow OSS CVE records to the CVE List referencing bulletins 7293015, 7293016 and 7294467.

    Sources:CVE-2026-93680 CVE record·CVE-2026-110186 CVE record·IBM Security Bulletin 7294467: Langflow OSS 1.0.0-1.12.8, fixed in 1.13.0

  6. Update

    HOL publishes operator guidance: upgrade path, auto-login lockout recovery, version and MCP endpoint checks, triage table and credential rotation.

    Source:HOL: Langflow CVE-2026-93680 and 32 more bugs, upgrade to 1.13.0

Publication clock: 0.7 hours from reviewed disclosure timestamp to HOL publication; 12-hour critical target; within target.

Reviewed artifacts

The specific artifacts this record concerns, as reviewed. Names and versions are shown as text; a registry link is not recorded here.

  • langflow@>=1.0.0, <=1.12.8 affected; fixed in 1.13.0 (subsets fixed in 1.12.4 and 1.12.5)pypipackage
  • langflowai/langflow@tags 1.0.0 through 1.12.8 affected; 1.13.0 fixeddockerpackage

Sources

Every claim on this record is traceable to the sources below.

Campaign questions, answered

On 2026-10-29 between 15:14 and 15:19 UTC, IBM (the CNA for Langflow) published 33 CVE records for Langflow OSS, backed by three IBM security bulletins dated 2026-10-19 (nodes 7293015 and 7293016) and 2026-10-28 (node 7294467). Four IDs affect every release from HOL Guard tracked the campaign from Oct 22, 2026 to Oct 29, 2026.

No. Active tracking ended Oct 29, 2026; this page remains as a labeled historical research record and does not imply the threat is currently active.

The record cites 13 primary sources, names 2 reviewed artifacts, and was last fully reviewed Oct 29, 2026 by HOL Guard Security Publishing. Its uncertainty and limitations are stated on the record itself.

Campaign guidance is defensive context. It does not auto-apply policy or claim universal campaign coverage.

The permalink above is permanent. The full record can be downloaded from this page as JSON, an artifact CSV, and a STIX 2.1 bundle, and corrections can be requested through the published correction path on the record. IBM published one-line descriptions with no endpoint names or proof of concept, and its 1.13.0 bulletin summary does not map its themes (incomplete auto-login remediation, MCP tool enumeration RCE, API key exposure, Agentic Assistant code…

Reviewed in full by HOL Guard Security Publishing.

Published
Last full review
Last modified

Report a correction

Record HGTC-2026-F73A68AD