LangBot 4.11.0 Missing Authenticity Check Allows Event Injection
Unauthenticated attackers can forge Lark events in LangBot versions up to 4.11.0 by bypassing signature verification, potentially driving automated workflows.
Key points
- LangBot versions through 4.11.0 lack authenticity verification for incoming Lark/Feishu events.
- Attackers with a bot UUID can inject fabricated message or card action events.
- The vulnerability allows driving LLM pipelines and approvals as other users without authentication.
LangBot versions up to 4.11.0 contain a high-severity missing authenticity verification flaw. Unauthenticated attackers can exploit this weakness to inject forged Lark or Feishu events, bypassing required signature checks.
What happened
The vulnerability, tracked as CVE-2026-112265 with a CVSS score of 8.3, stems from CWE-345 weaknesses in LangBot up to version 4.11.0. The issue allows unauthenticated attackers to POST fabricated events, specifically `im.message.receive_v1` or `card.action.trigger`, to the application.
Why it matters
Knowing a bot’s UUID is sufficient for an attacker to execute this injection. The forged events can drive the LLM pipeline, trigger tools, and execute approvals as other users. This bypasses standard authentication controls, potentially compromising workflow integrity.
Next steps
- Inventory all LangBot installations to identify versions at or below 4.11.0.
- Restrict network access to bot endpoints, limiting exposure to trusted networks only.
- Monitor for anomalous event volumes or unexpected workflow executions.
Detection notes
Security teams should examine application logs for unexpected POST requests to event endpoints. Look for high volumes of `im.message.receive_v1` or `card.action.trigger` events originating from unfamiliar IP addresses. Check whether signature validation failures are being logged or silently ignored. Containment usually involves temporarily blocking external access to the bot’s webhook URLs. Analyse workflow triggers for actions initiated without corresponding user activity in the frontend.
Verify that event payloads match expected formats and originate from known Lark or Feishu domains. Investigate any sudden changes in approval statuses or tool executions that lack clear user initiation. Review access logs for attempts to interact with bot UUIDs from unauthorised sources. Ensure that monitoring systems capture both successful and failed event submissions for forensic analysis.
What to do and how to stay safe: LangBot
- Restrict network access to LangBot endpoints to trusted IP ranges only.
- Monitor logs for unauthenticated POST requests to event ingestion URLs.
- Verify that all incoming Lark/Feishu events include valid signatures.
- Once the vendor provides an update, apply it immediately to affected systems.
Step-by-step guide: Emergency Patching: How to Fix Critical Flaws Without Breaking Production
General security guidance from the Packet Tribune newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
Which LangBot versions are affected by this vulnerability?
LangBot versions through 4.11.0 are affected by the missing authenticity verification flaw.
What specific events can attackers inject?
Attackers can inject forged `im.message.receive_v1` or `card.action.trigger` events.
What is the CVSS severity of CVE-2026-112265?
The vulnerability has a CVSS score of 8.3, rated as HIGH severity.
