CodeWhale: run_tests Tool Enables RCE via Malicious Repository Without Approval
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.
- Products
- Not specified
- Affected versions
- Not specified
- Fixed versions
- Not specified
- CVEs
- CVE-2026-45311
- Confirmed fixed CVEs
- Not specified
- Updated in this feed
- 7 Nov 2026, 20:28 UTC · Revision 1
Source evidence
CodeWhale: run_tests Tool Enables RCE via Malicious Repository Without Approval
core-osv · Original sourcePublished 28 May 2026, 17:32 UTC
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.
osv:CVE-2026-45311
Complete source record
{
"id": "ea3424f06b881b38f7839702e7e6d04ca16226b2c5c66bbe643bec44a5edf0e6",
"url": "https://osv.dev/vulnerability/CVE-2026-45311",
"method": "structured",
"payload": {
"kind": "advisory",
"title": "CodeWhale: run_tests Tool Enables RCE via Malicious Repository Without Approval",
"summary": "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.",
"evidence": "osv:CVE-2026-45311",
"products": [],
"severity": "critical",
"structured": {
"id": "CVE-2026-45311",
"aliases": [
"CVE-2026-45311",
"GHSA-wx44-2q6h-j6p8"
],
"metrics": [
{
"type": "CVSS_V3",
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H"
}
],
"affected": [
{
"ranges": [
{
"repo": "https://github.com/codewhale-hq/Codewhale",
"type": "GIT",
"events": [
{
"introduced": "3204f556af2cf5f0bcc604fb9fd230b3a5a2e965"
},
{
"fixed": "cfb5f08ae515f8dcf93aa10bdcf37dff3549e7ec"
}
],
"database_specific": {
"extracted_events": [
{
"range": [
{
"introduced": "0.3.0"
},
{
"fixed": "0.8.23"
}
],
"source": "AFFECTED_FIELD",
"strategy": "StringRangeExpression"
},
{
"range": [
{
"introduced": "0"
},
{
"fixed": "0.8.23"
}
],
"source": "DESCRIPTION"
}
]
}
}
],
"versions": [
"v0.8.22",
"v0.8.21",
"v0.8.20",
"v0.8.19",
"v0.8.18",
"v0.8.17",
"v0.8.15",
"v0.8.14",
"v0.8.13",
"v0.8.12",
"v0.8.11",
"v0.8.10",
"v0.8.9",
"v0.8.8",
"v0.8.7",
"v0.8.6",
"v0.8.5",
"v0.8.4",
"v0.8.3",
"v0.8.2",
"v0.8.1",
"v0.8.0",
"v0.7.9",
"v0.7.8",
"v0.7.7",
"v0.7.6",
"v0.7.5",
"v0.7.4",
"v0.7.2",
"v0.7.1",
"v0.7.0",
"v0.6.7",
"v0.6.6",
"v0.6.5",
"v0.6.4",
"v0.6.3",
"v0.6.2",
"v0.6.1",
"v0.5.2",
"v0.5.1",
"v0.5.0",
"v0.4.9",
"v0.4.8",
"v0.4.7",
"v0.4.6",
"v0.4.5",
"v0.4.4",
"v0.4.3",
"v0.4.2",
"v0.4.1",
"v0.4.0",
"v0.3.33",
"v0.3.32",
"v0.3.31",
"v0.3.28",
"v0.3.27",
"v0.3.26",
"v0.3.25",
"v0.3.24",
"v0.3.23",
"v0.3.22",
"v0.3.21",
"v0.3.20",
"v0.3.19",
"v0.3.18",
"v0.3.17",
"v0.3.16",
"v0.3.14",
"v0.3.13",
"v0.3.12",
"v0.3.11",
"v0.3.10",
"v0.3.5",
"v0.3.4",
"v0.3.3",
"v0.3.2",
"v0.3.1",
"v0.3.0"
],
"database_specific": {
"source": "https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-45311.json"
}
}
],
"modified": "2026-10-28T02:30:51.476915677Z",
"provider": "osv",
"references": [
"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/45xxx/CVE-2026-45311.json",
"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-wx44-2q6h-j6p8",
"https://nvd.nist.gov/vuln/detail/CVE-2026-45311"
],
"provider_metadata": {
"cwe_ids": [
"CWE-94"
],
"cna_assigner": "GitHub_M",
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/45xxx/CVE-2026-45311.json"
}
},
"related_cves": [
{
"value": "CVE-2026-45311",
"evidence": "/id"
}
],
"security_relevant": true,
"severity_evidence": "/severity/0/score"
},
"source_id": "core-osv",
"document_id": 520312,
"published_at": "2026-05-28T17:32:26.95+00:00"
}