CVE-2026-90938
HighCVSS 8.6Summary
LangBot's plugin runtime (pip package langbot_plugin) through version 0.4.17 starts a debug WebSocket server on 0.0.0.0:5401 with authentication based on a key that defaults to an empty string and is never set. A remote attacker can register an arbitrary 'debug plugin' without credentials, receive full context of IM messages, inject forged replies, send messages as bots, read knowledge bases, and register malicious tools.
Risk Assessment
The risk includes full compromise of IM communication, unauthorized data access, impersonation of bots, and persistent denial of service through malicious plugin registration.
Recommendation
It is recommended to immediately disable or secure port 5401, and set a strong plugin_debug_key. Monitor access to the WebSocket server and restrict it to trusted networks.
Other vulnerabilities in LangBot
See all- CVE-2026-90562High
LangBot before 4.10.11 generates password recovery keys with only 24 bits of entropy and applies no rate limiting to the unauthenticated reset-password endpoint. Remote attackers knowing the administrator email can exhaust the keyspace through concurrent requests to reset the admin password and gain account access.
- CVE-2026-54449High
LangBot version 4.10.7 and earlier has a vulnerability that allows any authenticated user to add or change an STDIO MCP server configuration without an adequate authorization boundary. An attacker can configure commands and arguments that are executed as a subprocess on the LangBot server, enabling arbitrary command execution with LangBot service privileges.
Original NVD description (English source)
LangBot's plugin runtime (pip package langbot_plugin) through 0.4.17 starts a debug WebSocket server on 0.0.0.0:5401 (/plugin/ws) whose authentication is gated on plugin_debug_key, which defaults to an empty string and is never set by the upstream repository, Docker image, or docker-compose (which additionally publishes port 5401 to the host); the key check is therefore skipped entirely. Any remote attacker able to reach the port can register an arbitrary "debug plugin" without credentials. Because events are broadcast to all initialized plugins without filtering, the attacker's plugin receives the full context of every IM message event (including private chats, message chains, and user/sender IDs in plaintext) and can inject forged replies, send messages as any configured bot, enumerate bot UUIDs, invoke configured LLM models, read knowledge-base contents, and register malicious tools that feed every user's LLM pipeline. Registering with "prod_mode": true causes later legitimate installations of a plugin with the same author/name to be rejected, resulting in persistent denial of service. No patched version was available at the time of publication.