{"api_version":"1","generated_at":"2026-05-30T03:12:39+00:00","cve":"CVE-2026-45374","urls":{"html":"https://cve.report/CVE-2026-45374","api":"https://cve.report/api/cve/CVE-2026-45374.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-45374","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-45374"},"summary":{"title":"CodeWhale: task_create Insecure Defaults Enable RCE via Prompt Injection in Project Files","description":"CodeWhale is a DeepSeek + MiMo coding agent in terminal. Prior to 0.8.26, the task_create tool spawns durable sub-agents that inherit two insecure defaults, allow_shell defaults to true (config.rs:1499: self.allow_shell.unwrap_or(true)) and auto_approve defaults to true (task_manager.rs:297: auto_approve: Some(true)). When a user approves a task_create call (which requires ApprovalRequirement::Required), they approve what appears to be a benign work prompt. However, the spawned sub-agent silently receives unrestricted, unapproved shell access. This vulnerability is fixed in 0.8.26.","state":"PUBLISHED","assigner":"GitHub_M","published_at":"2026-05-28 18:16:35","updated_at":"2026-05-28 18:40:37"},"problem_types":["CWE-94","CWE-94 CWE-94: Improper Control of Generation of Code ('Code Injection')"],"metrics":[{"version":"3.1","source":"security-advisories@github.com","type":"Secondary","score":"9.6","severity":"CRITICAL","vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H","data":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H","baseScore":9.6,"baseSeverity":"CRITICAL","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"REQUIRED","scope":"CHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH"}},{"version":"3.1","source":"CNA","type":"DECLARED","score":"9.6","severity":"CRITICAL","vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H","data":{"attackComplexity":"LOW","attackVector":"NETWORK","availabilityImpact":"HIGH","baseScore":9.6,"baseSeverity":"CRITICAL","confidentialityImpact":"HIGH","integrityImpact":"HIGH","privilegesRequired":"NONE","scope":"CHANGED","userInteraction":"REQUIRED","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H","version":"3.1"}}],"references":[{"url":"https://github.com/Hmbown/DeepSeek-TUI/releases/tag/v0.8.26","name":"https://github.com/Hmbown/DeepSeek-TUI/releases/tag/v0.8.26","refsource":"security-advisories@github.com","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-72w5-pf8h-xfp4","name":"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-72w5-pf8h-xfp4","refsource":"security-advisories@github.com","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-45374","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-45374","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Hmbown","product":"CodeWhale","version":"affected < 0.8.26","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":{"cve_year":"2026","cve_id":"45374","cve":"CVE-2026-45374","epss":"0.000410000","percentile":"0.129250000","score_date":"2026-05-29","updated_at":"2026-05-30 00:13:24"},"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"cna":{"affected":[{"product":"CodeWhale","vendor":"Hmbown","versions":[{"status":"affected","version":"< 0.8.26"}]}],"descriptions":[{"lang":"en","value":"CodeWhale is a DeepSeek + MiMo coding agent in terminal. Prior to 0.8.26, the task_create tool spawns durable sub-agents that inherit two insecure defaults, allow_shell defaults to true (config.rs:1499: self.allow_shell.unwrap_or(true)) and auto_approve defaults to true (task_manager.rs:297: auto_approve: Some(true)). When a user approves a task_create call (which requires ApprovalRequirement::Required), they approve what appears to be a benign work prompt. However, the spawned sub-agent silently receives unrestricted, unapproved shell access. This vulnerability is fixed in 0.8.26."}],"metrics":[{"cvssV3_1":{"attackComplexity":"LOW","attackVector":"NETWORK","availabilityImpact":"HIGH","baseScore":9.6,"baseSeverity":"CRITICAL","confidentialityImpact":"HIGH","integrityImpact":"HIGH","privilegesRequired":"NONE","scope":"CHANGED","userInteraction":"REQUIRED","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H","version":"3.1"}}],"problemTypes":[{"descriptions":[{"cweId":"CWE-94","description":"CWE-94: Improper Control of Generation of Code ('Code Injection')","lang":"en","type":"CWE"}]}],"providerMetadata":{"dateUpdated":"2026-05-28T17:29:11.005Z","orgId":"a0819718-46f1-4df5-94e2-005712e83aaa","shortName":"GitHub_M"},"references":[{"name":"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-72w5-pf8h-xfp4","tags":["x_refsource_CONFIRM"],"url":"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-72w5-pf8h-xfp4"},{"name":"https://github.com/Hmbown/DeepSeek-TUI/releases/tag/v0.8.26","tags":["x_refsource_MISC"],"url":"https://github.com/Hmbown/DeepSeek-TUI/releases/tag/v0.8.26"}],"source":{"advisory":"GHSA-72w5-pf8h-xfp4","discovery":"UNKNOWN"},"title":"CodeWhale: task_create Insecure Defaults Enable RCE via Prompt Injection in Project Files"}},"cveMetadata":{"assignerOrgId":"a0819718-46f1-4df5-94e2-005712e83aaa","assignerShortName":"GitHub_M","cveId":"CVE-2026-45374","datePublished":"2026-05-28T17:26:42.979Z","dateReserved":"2026-05-12T00:51:29.086Z","dateUpdated":"2026-05-28T17:29:11.005Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-05-28 18:16:35","lastModifiedDate":"2026-05-28 18:40:37","problem_types":["CWE-94","CWE-94 CWE-94: Improper Control of Generation of Code ('Code Injection')"],"metrics":{"cvssMetricV31":[{"source":"security-advisories@github.com","type":"Secondary","cvssData":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H","baseScore":9.6,"baseSeverity":"CRITICAL","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"REQUIRED","scope":"CHANGED","confidentialityImpact":"HIGH","integrityImpact":"HIGH","availabilityImpact":"HIGH"},"exploitabilityScore":2.8,"impactScore":6}]},"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"45374","Ordinal":"1","Title":"CodeWhale: task_create Insecure Defaults Enable RCE via Prompt I","CVE":"CVE-2026-45374","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"45374","Ordinal":"1","NoteData":"CodeWhale is a DeepSeek + MiMo coding agent in terminal. Prior to 0.8.26, the task_create tool spawns durable sub-agents that inherit two insecure defaults, allow_shell defaults to true (config.rs:1499: self.allow_shell.unwrap_or(true)) and auto_approve defaults to true (task_manager.rs:297: auto_approve: Some(true)). When a user approves a task_create call (which requires ApprovalRequirement::Required), they approve what appears to be a benign work prompt. However, the spawned sub-agent silently receives unrestricted, unapproved shell access. This vulnerability is fixed in 0.8.26.","Type":"Description","Title":"CodeWhale: task_create Insecure Defaults Enable RCE via Prompt I"}]}}}