LangBot 4.11.0 Flaw Lets Attackers Forge Lark Events Without Auth

LangBot versions through 4.11.0 contain a high-severity vulnerability that allows unauthenticated attackers to inject forged events. The flaw bypasses signature checks for Lark and Feishu integrations, enabling unauthorized manipulation of bot workflows.

What happened

The National Vulnerability Database records CVE-2026-112265 with a CVSS score of 8.3. The issue is classified as CWE-345, indicating missing authenticity verification. According to the NVD record, LangBot versions up to 4.11.0 fail to properly verify the authenticity of incoming requests.

An attacker who knows a bot UUID can exploit this gap. They can POST fabricated events, such as im.message.receive_v1 or card.action.trigger, directly to the system. This action skips necessary signature checks that should validate the source of the data.

Why it matters

Successful exploitation allows an attacker to drive the LLM pipeline, tools, and approvals. The NVD description states these actions can be performed as other users. This means an unauthenticated actor can manipulate bot behavior without valid credentials.

The vulnerability affects the core integration layer for Lark and Feishu. Because signature checks are skipped, the system accepts forged events as legitimate. This could lead to unauthorized data processing or approval actions within the connected enterprise applications.

What to watch

• Monitor for unexpected LLM pipeline activity or approval requests originating from known bot UUIDs.
• Review logs for POST requests containing im.message.receive_v1 or card.action.trigger events.
• Check if any LangBot instances are running versions 4.11.0 or earlier in your environment.
• Verify that all incoming Lark or Feishu events are being validated for authenticity before processing.

Who needs to act first

Teams should prioritize updating exposed LangBot instances immediately. Since no fix is confirmed, isolate systems that accept unauthenticated external requests. Verify that only trusted internal networks can reach the bot endpoints temporarily. Back up configuration files before making any changes. Test any mitigation steps in a non-production environment first to ensure business continuity.

Background: CVSS score

The CVSS score explains vulnerability severity using a 0.0 to 10.0 scale. It measures attack vector, complexity, privileges required, and impact on confidentiality, integrity, and availability. You use this metric to prioritize patches, though you must adjust it for your specific environment to avoid wasting time on irrelevant threats.

Read the full guide: CVSS Score Explained: How to Read Risk Metrics and Prioritize Fixes

What to do and how to stay safe: LangBot

• Inventory all LangBot installations and identify those running version 4.11.0 or earlier.
• Restrict network access to LangBot endpoints to only trusted sources until a patch is available.
• Monitor authentication logs for unauthorized POST requests to bot endpoints.
• Contact the vendor to confirm if a security update is being developed for this vulnerability.

Step-by-step guide: Virtual Patching Myths: What It Actually Fixes and Breaks

General security guidance from the Patch Gazette newsroom. It is not confirmed advice from the organisations named in this story.

Frequently asked questions

Which versions of LangBot are affected by CVE-2026-112265?

LangBot versions up to and including 4.11.0 are affected by this vulnerability.

Is there a patch available for this vulnerability yet?

No, the source material does not confirm that a fix or patch has been released.

What can an attacker do with this vulnerability?

An attacker can inject forged Lark/Feishu events to drive LLM pipelines and approvals as other users.