Langflow /api/v1/responses IDOR — cross-user flow execution
An authenticated IDOR in Langflow's /api/v1/responses endpoint lets a logged-in attacker execute any other user's flow by passing the victim's flow UUID. NVD scores it 8.4 high; the vendor GHSA calls it 9.9 critical. Fixed in Langflow 1.9.1. Added to CISA KEV on 2026-07-07.
- Vendor
- Langflow
- Product
- Langflow (versions before 1.9.1)
- CVSS
- 8.4
- EPSS (exploit probability)
- 0.9%
- Status
- kev
- CISA patch-by (BOD 22-01)
- Published
Is CVE-2026-55255 being exploited?
Yes. CISA added CVE-2026-55255 to its Known Exploited Vulnerabilities catalog on . CISA only lists a CVE when it has reliable evidence of exploitation in the wild.
FIRST's EPSS model puts the probability of exploitation activity in the next 30 days at 0.9% (scored October 14, 2026).
- CISA KEV[ LISTED ]
- ENISA EUVD (exploited)[ — ]
- CIRCL KEV[ — ]
Catalog snapshot 2026-11-22. Full cross-catalog view: /kev-comparison
CVE-2026-55255 timeline
Every event links to its source. Catalog dates are each catalog's own; day offsets count from NVD publication. Data: JSON
What to do
US federal civilian agencies must remediate by (3 days after listing) under BOD 22-01. Everyone else: that deadline is a sensible upper bound, not a target. Vendor guidance and the NVD record are linked under Source above.
CVE-2026-55255 is an insecure direct object reference (IDOR) in Langflow’s /api/v1/responses endpoint. Per the vendor’s GHSA-qrpv-q767-xqq2, the get_flow_by_id_or_endpoint_name helper validated ownership on endpoint-name lookups but not on UUID lookups — so an authenticated user could execute any other user’s flow by specifying the victim’s flow UUID in the request. Fixed in Langflow 1.9.1 (PR #12832), which enforces ownership validation on both lookup branches and returns 404 for cross-user access rather than leaking flow existence via error codes.
Severity note
NVD scores CVE-2026-55255 at 8.4 (high). Langflow’s own advisory calls it 9.9 (critical). This site defers to NVD for the severity badge and flags the vendor’s framing in prose — same policy applied on BeyondTrust’s four-CVE RS/PRA bundle three days ago. The gap between the two scores turns on assumed impact scope; a shared-team Langflow deployment where workflow-editor accounts are handed out liberally sits closer to the vendor’s score than NVD’s.
Advisory context
CISA added CVE-2026-55255 to the Known Exploited Vulnerabilities catalog on 2026-07-07 alongside two Joomla page-builder RCEs — CVE-2026-56290 (Joomlack Page Builder) and CVE-2026-48908 (JoomShaper SP Page Builder). Federal due date: 2026-07-10.
This is the second Langflow entry on KEV in two weeks — a separate Langflow code-execution flaw was the initial-access vector in Sysdig’s JADEPUFFER ransomware writeup covered earlier this month.
What to do
Upgrade to Langflow 1.9.1. If a same-week upgrade is not possible, restrict /api/v1/responses at the ingress and audit the last 30 days of flow-execution logs for cross-user activity — see the article writeup for the full priority order across the three July 7 KEV adds.
NVD’s canonical record: CVE-2026-55255.
Other Langflow CVEs we track
| CVE / Vulnerability | Severity | CVSS | EPSS | Added to KEV |
|---|---|---|---|---|
| CVE-2026-93674 Langflow Authenticated Blind Command Injectionwrite-up | [ CRITICAL ] | 9.8 | 14.1% | — |
| CVE-2026-7873 Langflow Authenticated RCE via Code Validation Endpointwrite-up | [ CRITICAL ] | 9.9 | 6.9% | — |
| CVE-2026-10134 Langflow Unauthenticated RCE in Public Flow Build Endpointwrite-up | [ CRITICAL ] | 10.0 | 2.3% | — |
| CVE-2026-0769 Langflow Unauthenticated RCE via Custom Component Evaluationwrite-up | [ CRITICAL ] | 9.8 | 41.2% | — |
| CVE-2026-105740 Authenticated RCE via MCP tool additionwrite-up | [ CRITICAL ] | 9.9 | 3.9% | — |
| CVE-2026-0768 Langflow Unauthenticated RCE via Code Execution Endpointwrite-up | [ CRITICAL ] | 9.8 | 28.1% | — |
| CVE-2026-0770 Langflow validate-endpoint exec_globals RCE (unauthenticated)write-up | [ CRITICAL ] | 9.8 | 62.6% | |
| CVE-2025-34291 Langflow CORS misconfiguration enables CSRF-driven token theft and RCEwrite-up | [ HIGH ] | 8.8 | 93.2% |
