---
id: CVE-2026-75858
title: >-
  CodeWhale (packages codewhale / codewhale-tui) versions >= 0.8.41 and < 0.8.64
  contain a remote code execution vulnerability in the rlm_eval tool
summary: >-
  CodeWhale (packages codewhale / codewhale-tui) versions >= 0.8.41 and < 0.8.64
  contain a remote code execution vulnerability in the rlm_eval tool. The tool's
  approval_requirement() returns ApprovalRequirement::Auto, which the engine
  trea…
severity: high
cvss: 7.8
cvssVector: 'CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H'
cwe:
  - CWE-94
  - CWE-862
vendor: deepseek-tui
product: deepseek-tui
affected:
  - 'deepseek-tui >= 0.8.33, <= 0.8.41'
  - 'deepseek-tui >= 0.8.33, < 0.8.41'
  - 'codewhale-tui >= 0.8.41, < 0.8.64'
  - 'codewhale >= 0.8.41, < 0.8.64'
patched:
  - deepseek-tui 0.8.41
  - codewhale-tui 0.8.64
  - codewhale 0.8.64
published: '2026-08-18'
updated: '2026-09-08'
sourceUpdated: '2026-09-08T20:32:39.347'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-75858'
references:
  - url: >-
      https://github.com/Hmbown/CodeWhale/commit/57f3c89471e27ac4032d9791f6885e5d4408c381
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-wrj3-vj8c-784f
    label: disclosure@vulncheck.com
  - url: >-
      https://www.vulncheck.com/advisories/codewhale-rlm-eval-before-remote-code-execution
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-wrj3-vj8c-784f
    label: 134c704f-9b21-4f2e-91b3-4a467353bcc0
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-75858'
  - url: 'https://github.com/advisories/GHSA-wrj3-vj8c-784f'
tags:
  - nvd
  - ghsa
  - rust
epss: 0.00364
epssPercentile: 0.27507
aliases:
  - GHSA-wrj3-vj8c-784f
ecosystem: rust
ingestedAt: '2026-09-04T18:25:56.981Z'
---

## Overview

CodeWhale (packages codewhale / codewhale-tui) versions >= 0.8.41 and < 0.8.64 contain a remote code execution vulnerability in the rlm_eval tool. The tool's approval_requirement() returns ApprovalRequirement::Auto, which the engine treats as 'never prompt,' causing arbitrary model-supplied Python code to run in a python3 interpreter without consulting the user's configured --approval-policy and without any approval prompt or audit step. An attacker can induce the agent to execute arbitrary code via prompt injection in untrusted content the agent reads (a web page, fetched URL, repository file, or MCP tool result); the companion rlm_open tool can stage such content. Code runs on the user's machine at the user's privilege level. Fixed in 0.8.64.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2026-75858)

Affected packages:

- `deepseek-tui >= 0.8.33, <= 0.8.41`
- `deepseek-tui >= 0.8.33, < 0.8.41`
- `codewhale-tui >= 0.8.41, < 0.8.64`
- `codewhale >= 0.8.41, < 0.8.64`

Patched in:

- `deepseek-tui 0.8.41`
- `codewhale-tui 0.8.64`
- `codewhale 0.8.64`

Source: https://github.com/advisories/GHSA-wrj3-vj8c-784f