Skip to content

CVE-2026-75858 CodeWhale rlm_eval before 0.8.64 Remote Code Execution

HighCVSS 8.5 EPSS 0.36%
Published
Updated
Exploit status
No public exploit known
CISA SSVC
Proof of concept public · automatable: no · technical impact: total

Description

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.

Upgrade path

8 packages: 8 fixed.

PackageAffectedFixed inStatus
crates.io codewhale-tui >= 0.8.41, < 0.8.64 0.8.64 Fixed
crates.io codewhale-tui(crates.io) >= 0.8.41, < 0.8.64 0.8.64 Fixed
crates.io deepseek-tui >= 0.8.33, < 0.8.41 0.8.41 Fixed
crates.io deepseek-tui(crates.io) >= 0.8.33, <= 0.8.41 after 0.8.41 Fixed
npm codewhale >= 0.8.41, < 0.8.64 0.8.64 Fixed
npm codewhale(npm) >= 0.8.41, < 0.8.64 0.8.64 Fixed
npm deepseek-tui >= 0.8.33, < 0.8.41 0.8.41 Fixed
npm deepseek-tui(npm) >= 0.8.33, < 0.8.41 0.8.41 Fixed

Affected products

ProductAffected versionsFixed in
crates.io deepseek-tui(crates.io) >= 0.8.33, <= 0.8.41 No fix listed
npm deepseek-tui(npm) >= 0.8.33, < 0.8.41 0.8.41
crates.io codewhale-tui(crates.io) >= 0.8.41, < 0.8.64 0.8.64
npm codewhale(npm) >= 0.8.41, < 0.8.64 0.8.64
crates.io deepseek-tui >= 0.8.33, < 0.8.41 0.8.41
npm deepseek-tui >= 0.8.33, < 0.8.41 0.8.41
crates.io codewhale-tui >= 0.8.41, < 0.8.64 0.8.64
npm codewhale >= 0.8.41, < 0.8.64 0.8.64

References