9.6
/ 10
CRITICAL
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
Description
CodeWhale is a DeepSeek + MiMo coding agent in terminal. From 0.3.0 to 0.8.23, the run_tests tool executes cargo test in the workspace with ApprovalRequirement::Auto, meaning it runs without any user approval prompt. cargo test compiles and executes arbitrary code: test binaries, build.rs build scripts, and proc macros. While auto-approving test execution is a deliberate design choice, it creates an inconsistency in the security boundary. However, in a malicious repository, test code can execute arbitrary shell commands, exfiltrate credentials, or establish persistence with zero approval. The attack is amplified by AGENTS.md (auto-loaded into the system prompt), which can instruct the model to run tests proactively at session start. This vulnerability is fixed in 0.8.23.
AI Analysis
Remote Code Execution (RCE) vulnerability in CodeWhale due to the run_tests tool executing cargo test without user approval, allowing malicious repositories to execute arbitrary shell commands, exfiltrate credentials, or establish persistence.
Basic Information
ID
CVE-2026-45311
Source
GitHub_M
Published
May 28, 2026 at 17:32
Affected Product
Vendor
Hmbown
Product
CodeWhale
Version
>= 0.3.0, < 0.8.23
Affected Versions
Hmbown CodeWhale >= 0.3.0, < 0.8.23
CWE Classification
AI Assessment
AI Score
9.6 / 10
AI Severity
Critical
Vendor
Hmbown
Product
CodeWhale
Version
0.3.0-0.8.23