Arbitrary Command Execution in AnythingLLM by Mintplex Labs
CVE-2026-48116

7.5HIGH

Key Information:

Vendor

Mintplex-labs

Status
Anything-llm
Vendor
CVE Published:
28 May 2026

What is CVE-2026-48116?

The AnythingLLM application, which enhances LLM interactions by allowing content references during chats, is vulnerable to arbitrary command execution. Versions prior to 1.13.0 suffer from an issue where the filesystem-search-files agent skill improperly passes a pattern parameter to ripgrep without a proper end-of-options separator. As a result, an attacker who engages with an agent in a deployment with the filesystem plugin enabled can exploit this to execute arbitrary commands inside the AnythingLLM server container. This critical flaw can allow attackers to manipulate the server environment by running any command, compromising the security and integrity of the application. The vulnerability has been addressed in version 1.13.0.

Affected Version(s)

anything-llm < 1.13.0

References

https://github.com/Mintplex-Labs/anything-llm/security/ad...
x_refsource_CONFIRM
https://github.com/Mintplex-Labs/anything-llm/commit/94ed...
x_refsource_MISC

CVSS V3.1

Score:
7.5
Severity:
HIGH
Confidentiality:
High
Integrity:
High
Availability:
High
Attack Vector:
Network
Attack Complexity:
High
Privileges Required:
Low
User Interaction:
None
Scope:
Unchanged

Timeline

  • Vulnerability published

  • Vulnerability Reserved

.