{"id":"CVE-2026-18733","aliases":["EUVD-2026-52432","GHSA-mqvc-p852-wf8x"],"status":"published","title":"Prompt injection bypasses shell tool consent gate in Strands Agents Tools","description":"A prompt injection vulnerability in the shell tool in Amazon Strands Agents Tools before 0.8.0 might allow remote actors to execute arbitrary operating system commands on the agent's host via a crafted prompt that sets the non_interactive parameter to true, bypassing the human consent gate.\n\n\n\nTo remediate this issue, users should upgrade to version 0.8.0.","severity":{"version":"4.0","score":7.5,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"high","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-11","score":0.00777,"percentile":0.56263,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":false},"cwes":["CWE-1427"],"published":"2026-08-03T20:33:24Z","modified":"2026-10-28T02:30:55Z","products":[{"vendor":"amazon","product":"strands-agents-tools"},{"vendor":"aws","product":"strands-agents-tools"}],"references":[{"url":"https://pypi.org/project/strands-agents-tools/0.8.0/","tags":["release-notes","patch","fix"]},{"url":"https://aws.amazon.com/security/security-bulletins/2026-072-aws/","tags":["vendor-advisory","advisory"]},{"url":"https://github.com/strands-agents/tools/security/advisories/GHSA-mqvc-p852-wf8x","tags":["release-notes","advisory"]},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/18xxx/CVE-2026-18733.json","tags":["advisory"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-18733","tags":["advisory"]}],"sources":{"cve":[{"id":"CVE-2026-18733","title":"Prompt injection bypasses shell tool consent gate in Strands Agents Tools","description":"A prompt injection vulnerability in the shell tool in Amazon Strands Agents Tools before 0.8.0 might allow remote actors to execute arbitrary operating system commands on the agent's host via a crafted prompt that sets the non_interactive parameter to true, bypassing the human consent gate.\n\n\n\nTo remediate this issue, users should upgrade to version 0.8.0.","status":"published","published":"2026-08-03T20:33:24.872Z","modified":"2026-08-04T19:31:53.415Z","cvss":[{"version":"4.0","score":7.5,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"high","assessor":"cna"},{"version":"3.1","score":8.8,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","level":"high","assessor":"cna"}],"cwes":["CWE-1427"],"updated_at":"2026-11-02T14:20:12.260247Z"}],"euvd":[{"id":"EUVD-2026-52432","description":"A prompt injection vulnerability in the shell tool in Amazon Strands Agents Tools before 0.8.0 might allow remote actors to execute arbitrary operating system commands on the agent's host via a crafted prompt that sets the non_interactive parameter to true, bypassing the human consent gate.\n\n\n\nTo remediate this issue, users should upgrade to version 0.8.0.","status":"published","published":"2026-08-03T20:33:24Z","modified":"2026-08-04T19:31:53Z","cvss":[{"version":"4.0","score":7.5,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"high"}],"updated_at":"2026-11-02T18:02:22.516769Z"}],"nvd":[{"id":"CVE-2026-18733","description":"A prompt injection vulnerability in the shell tool in Amazon Strands Agents Tools before 0.8.0 might allow remote actors to execute arbitrary operating system commands on the agent's host via a crafted prompt that sets the non_interactive parameter to true, bypassing the human consent gate.\n\n\n\nTo remediate this issue, users should upgrade to version 0.8.0.","status":"published","published":"2026-08-03T21:16:37.963Z","modified":"2026-10-23T21:10:00.157Z","cvss":[{"version":"4.0","score":7.5,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X","level":"high","assessor":"secondary"},{"version":"3.1","score":8.8,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","level":"high","assessor":"secondary"}],"updated_at":"2026-11-02T14:12:34.643497Z"}],"osv":[{"id":"CVE-2026-18733","title":"Prompt injection bypasses shell tool consent gate in Strands Agents Tools","description":"A prompt injection vulnerability in the shell tool in Amazon Strands Agents Tools before 0.8.0 might allow remote actors to execute arbitrary operating system commands on the agent's host via a crafted prompt that sets the non_interactive parameter to true, bypassing the human consent gate.\n\n\n\nTo remediate this issue, users should upgrade to version 0.8.0.","status":"published","published":"2026-08-03T20:33:24.872Z","modified":"2026-10-28T02:30:55.880266918Z","cvss":[{"version":"4.0","score":7.5,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"high"}],"cwes":["CWE-1427"],"updated_at":"2026-11-02T15:44:58.024074Z"}]}}