PatchSiren

PatchSiren cyber security CVE debrief

CVE-2026-72917 Mintplex-Labs CVE debrief

An attacker can exploit a vulnerability in AnythingLLM's account-recovery flow to take over a user account, including an administrator account, by submitting a valid recovery code with different surrounding whitespace. The vulnerability exists in versions 1.0.0 to 1.15.0 of the application. This issue arises from the unauthenticated account-recovery flow in server/utils/PasswordRecovery/index.js, where the recoverAccount() function is used to deduplicate raw recoveryCodes values before trimming them, allowing an attacker to satisfy the two-code check with a valid recovery code that has different surrounding whitespace.

Vendor
Mintplex-Labs
Product
anything-llm
CVSS
MEDIUM 5.9
CISA KEV
Not listed in stored evidence
Original CVE published
2026-08-10
Original CVE updated
2026-09-18
Advisory published
2026-08-10
Advisory updated
2026-09-18

Who should care

Defenders responsible for deploying and managing AnythingLLM instances, especially in multi-user environments, should assess their exposure and prioritize remediation. This includes IT security teams, system administrators, and compliance officers who oversee the security and integrity of user accounts and sensitive data.

Why it matters

CVE-2026-72917 is a vulnerability in AnythingLLM's account-recovery flow that allows an attacker to take over a user account, including an administrator account, by submitting a valid recovery code with different surrounding whitespace. Defenders responsible for deploying and managing AnythingLLM instances should assess their exposure and prioritize remediation.

  • An attacker can take over a user account, including an administrator account, by exploiting this vulnerability.
  • The vulnerability allows for unauthorized access to user accounts without requiring authentication.
  • Defenders need to verify and remediate this vulnerability to prevent potential account takeovers.
  • The vulnerability requires verification of the AnythingLLM version and implementation of compensating controls to prevent exploitation.

Technical summary

The vulnerability exists in the unauthenticated account-recovery flow of AnythingLLM, specifically in the server/utils/PasswordRecovery/index.js file. The recoverAccount() function is used to deduplicate raw recoveryCodes values before trimming them, allowing an attacker to submit a valid recovery code with different surrounding whitespace to satisfy the two-code check. This issue enables an attacker who knows the target username and one recovery code to call POST /api/system/recover-account in multi-user mode, receive a password-reset token, and use POST /api/system/reset-password to take over the account.

Defensive priority

Defenders should prioritize verifying and remediating this vulnerability, especially for multi-user deployments of AnythingLLM.

Recommended defensive actions

  • Verify the version of AnythingLLM and check if it is within the affected range (1.0.0 to 1.15.0).
  • Implement authentication for the account-recovery flow.
  • Restrict access to the account-recovery endpoint.
  • Monitor for suspicious activity on the account-recovery endpoint.
  • Perform a thorough review of the system's logs to identify any potential exploitation attempts.
  • Update the AnythingLLM application to a version outside of the affected range.
  • Review and update incident response plans to include procedures for handling potential account takeovers.

Evidence notes

The vulnerability is caused by the unauthenticated account-recovery flow in server/utils/PasswordRecovery/index.js, which uses recoverAccount() to deduplicate raw recoveryCodes values before trimming them.

Sources and references

Verified primary and authoritative sources

  • CVE-2026-72917 CVE Program record

    Publisher, destination, and source semantics verified

    URL: https://www.cve.org/CVERecord?id=CVE-2026-72917

    CVE Program - Official CVE Program record with source-provided CVE metadata.

  • CVE-2026-72917 NVD vulnerability detail

    Publisher, destination, and source semantics verified

    URL: https://nvd.nist.gov/vuln/detail/CVE-2026-72917

    NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.

Supplemental references

Methodology and review provenance

AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.