CVE-2026-45311CWE-94

CodeWhale: run_tests Tool Enables RCE via Malicious Repository Without Approval

Critical · published May 28, 2026

CVSS v3.1
9.6
EPSS
1%
Percentile
52.0
In the wild
Unconfirmed
What it is

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.

The record
Technical detail
CVSS v3.1
9.6 · CRITICAL
Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
CVSS v4.0
Not supplied
EPSS
0.00685 · 52.0th percentile
Weakness
CWE-94 · Improper Control of Generation of Code ('Code Injection')
Published
2026-05-28T17:32Z
EPSS history
Timeline
  • 28 MAY 17:32Z
    CodeWhale: run_tests Tool Enables RCE via Malicious Repository Without Approval
    cvelistv5