{"id":"EUVD-2026-60965","enisaUuid":"7d18f7ed-fdaf-39ee-ad22-2191370eeeae","description":"CodeWhale (packages codewhale / codewhale-tui) versions \u003e\u003d 0.8.41 and \u003c 0.8.64 contain a remote code execution vulnerability in the rlm_eval tool. The tool\u0027s approval_requirement() returns ApprovalRequirement::Auto, which the engine treats as \u0027never prompt,\u0027 causing arbitrary model-supplied Python code to run in a python3 interpreter without consulting the user\u0027s 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\u0027s machine at the user\u0027s privilege level. Fixed in 0.8.64.","datePublished":"Aug 18, 2026, 3:21:57 PM","dateUpdated":"Aug 18, 2026, 5:59:11 PM","baseScore":8.5,"baseScoreVersion":"4.0","baseScoreVector":"CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","references":"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-wrj3-vj8c-784f\nhttps://github.com/Hmbown/CodeWhale/commit/57f3c89471e27ac4032d9791f6885e5d4408c381\nhttps://www.vulncheck.com/advisories/codewhale-rlm-eval-before-remote-code-execution\n","aliases":"CVE-2026-75858\nGHSA-wrj3-vj8c-784f\n","assigner":"VulnCheck","epss":0.36,"enisaIdProduct":[{"id":"18c37042-7ecf-3610-b412-7e47fe4b3021","product":{"name":"CodeWhale","vendor":{"name":"Hmbown"}},"product_version":"0.8.33 \u003c0.8.41"},{"id":"c91167e0-d41a-3b2c-a126-a4db69ad447d","product":{"name":"CodeWhale","vendor":{"name":"Hmbown"}},"product_version":"0.8.41 \u003c0.8.64"}],"enisaIdVendor":[{"id":"b9a5039f-ac0d-341d-87e0-1eafec3e27bb","vendor":{"name":"Hmbown"}}],"enisaIdVulnerability":[{"id":"81d609b6-aeb9-34fd-b4b9-abf660b8a449","vulnerability":{"id":"GHSA-wrj3-vj8c-784f","description":"CodeWhale: rlm_eval auto-approves arbitrary Python execution, bypassing the user\u0027s approval policy (RCE)","datePublished":"Sep 4, 2026, 6:12:15 PM","dateUpdated":"Sep 4, 2026, 6:12:15 PM","baseScore":0.0,"references":"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-wrj3-vj8c-784f\nhttps://nvd.nist.gov/vuln/detail/CVE-2026-75858\nhttps://github.com/Hmbown/CodeWhale/commit/57f3c89471e27ac4032d9791f6885e5d4408c381\nhttps://github.com/Hmbown/CodeWhale\nhttps://www.vulncheck.com/advisories/codewhale-rlm-eval-before-remote-code-execution\n","enisa_id":"EUVD-2026-60965\n","aliases":"GHSA-wrj3-vj8c-784f\nCVE-2026-75858\n","epss":0.0,"dataProcessed":"Nov 1, 2026, 2:05:08 AM","vulnerabilityProduct":[],"vulnerabilityVendor":[]}},{"id":"ddc58dd4-b00f-3739-9f9e-aab7621aafa2","vulnerability":{"id":"CVE-2026-75858","description":"CodeWhale (packages codewhale / codewhale-tui) versions \u003e\u003d 0.8.41 and \u003c 0.8.64 contain a remote code execution vulnerability in the rlm_eval tool. The tool\u0027s approval_requirement() returns ApprovalRequirement::Auto, which the engine treats as \u0027never prompt,\u0027 causing arbitrary model-supplied Python code to run in a python3 interpreter without consulting the user\u0027s 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\u0027s machine at the user\u0027s privilege level. Fixed in 0.8.64.","datePublished":"Aug 18, 2026, 3:21:57 PM","dateUpdated":"Aug 18, 2026, 5:59:11 PM","status":"PUBLISHED","baseScore":8.5,"baseScoreVersion":"4.0","baseScoreVector":"CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","references":"https://github.com/Hmbown/CodeWhale/security/advisories/GHSA-wrj3-vj8c-784f\nhttps://github.com/Hmbown/CodeWhale/commit/57f3c89471e27ac4032d9791f6885e5d4408c381\nhttps://www.vulncheck.com/advisories/codewhale-rlm-eval-before-remote-code-execution\n","enisa_id":"EUVD-2026-60965\n","assigner":"VulnCheck","epss":0.36,"dataProcessed":"Nov 3, 2026, 7:15:15 AM","vulnerabilityProduct":[{"id":"0ee58515-88cc-36d4-9e58-9fdfa3089910","product":{"name":"CodeWhale","vendor":{"name":"Hmbown"}},"product_version":"0.8.41 \u003c0.8.64"},{"id":"c4812c26-7df9-39c7-9d6a-a096f45944fe","product":{"name":"CodeWhale","vendor":{"name":"Hmbown"}},"product_version":"0.8.33 \u003c0.8.41"}],"vulnerabilityVendor":[{"id":"5fc5b85d-bd64-3c8f-b440-150292158d63","vendor":{"name":"Hmbown"}}]}}],"enisaIdAdvisory":[]}