// Security Advisory Triage
Langflow prompt injection enables code execution, patched in 1.10.3
GitHub advisories — pip · published 2026-10-06 18:26 UTC
WHAT CHANGEDGitHub published CVE-2026-7700 on 5 October 2026, a high-severity prompt injection vulnerability in Langflow's Smart Transform component that allows code execution. The fix shipped in version 1.10.3.
WHY IT MATTERSAn attacker who can control prompts processed by Smart Transform can execute arbitrary code in your Langflow runtime. This is a direct path from user input to system compromise in any workflow using that component.
WHAT TO DOUpgrade langflow to 1.10.3 or later. If you cannot upgrade immediately, audit every workflow for Smart Transform usage and disable or isolate those workflows until you patch.
Model-reported confidence 95% (not measured accuracy)
Primary source ↗
written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
+ghsa_id: GHSA-9fpm-3445-2vx4 +summary: Langflow: Prompt injection in Langflow Smart Transform can lead to code execution +severity: high +published_at: 2026-10-05T22:30:41Z +cve_id: CVE-2026-7700 +vulnerabilities[].package.name: langflow +vulnerabilities[].first_patched_version: 1.10.3
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
| triage verdict | MATERIAL at 0.92, needed 0.5 |
|---|---|
| triage reasoning | Langflow is an AI agent framework in dependency trees. Prompt injection leading to code execution is exploitable and actionable. |
| signals | AI framework SDK affected (Langflow) · prompt injection vulnerability · code execution impact · high severity rating · patched version available (1.10.3) · new advisory entry |
| suppression | — |
| triage model | anthropic/claude-haiku-4.5 ($0.00000) |
| writer model | anthropic/claude-sonnet-4.5 ($0.00000) |