CVE-2026-109440

LOWCVSS 8.6 · HIGH

Immediate actions

Track advisory updates for patch or workaround availability

Recommended action window: Monitor and triage in normal cycle

NVD description

LangAlpha through 2026.09.07 contains an indirect prompt injection vulnerability that allows remote attackers to execute shell commands or Python by planting instructions in web content retrieved by the agent. Because Bash and ExecuteCode lack an allow-list and human-in-the-loop approval, attackers can run code in the workspace sandbox and exfiltrate vault secrets from _internal/.vault_secrets.json.

Back to dashboard

0.0/ 10 priority

Sources & remediation

Other references

Priority

LOW

Exploitation

NONE

PoC

NONE

Patch

NONE

Momentum

NONE

Deep dive

Full discourse1 post

Harsha Hegde@HarshaHegde021h
All 13 👇 DB-GPT: CVE-2026-77932 (9.8), CVE-2026-77933 (9.8) Letta: CVE-2026-77936 (9.8), CVE-2026-77937 Langchain-Chatchat: CVE-2026-75128 (9.8) piflow: CVE-2026-109442 (9.8) Coze Studio: CVE-2026-77934, CVE-2026-77935 Aix-DB: CVE-2026-77926 OpenSRE: CVE-2026-109444 JoySafeter: CVE-2026-109443 EvoScientist: CVE-2026-77930 LangAlpha: CVE-2026-109440 only coze studio has a fix so far (0.5.0) Full breakdown: https://severitydaily.com/ai-agent-frameworks-thirteen-records-vulncheck-lasso-cwe-306-cwe-1188-shipped-defaults-no-fixed-version/
0000056
14 followersView on X

Explore more

Back to dashboard