{"id":"CVE-2026-110248","kind":"vulnerability","aliases":["CVE-2026-110248"],"related":[],"title":"Amazon SageMaker Python SDK: failed Git clones log repository credentials in plaintext","description":"CVE-2026-110248 - Insertion of sensitive information into log files in Amazon SageMaker Python SDK\n\n### CVE-2026-110248 - Insertion of sensitive information into log files in Amazon SageMaker Python SDK\n\nBulletin ID: 2026-134-AWS Scope: AWS Content Type: Important (requires attention) Publication Date: 10/14/2026 11:30 AM PDT\n\nDescription:\n\nAmazon SageMaker Python SDK is an open source library for training and deploying machine learning models on Amazon SageMaker.\n\nWe identified CVE-2026-110248, an issue in the git_utils.py component of Amazon SageMaker Python SDK. When a Git clone operation fails, repository credentials in subprocess error output are written to CloudWatch Logs without redaction. An authenticated remote user with access to those logs is able to obtain the plaintext credentials.\n\nImpacted versions:\n\n- Amazon SageMaker Python SDK V2 versions before 2.257.3\n- Amazon SageMaker Python SDK V3 versions before 3.10.0\n\nResolution:\n\nThis issue has been addressed in Amazon SageMaker Python SDK V2 version 2.257.3 and V3 version 3.10.0. We recommend upgrading to the latest version and ensuring any forked or derivative code incorporates the fixes.\n\nWorkarounds:\n\nIf immediate upgrading is not possible, avoid including credentials in Git repository URLs and restrict access to CloudWatch Logs containing SageMaker Python SDK output. Review logs from failed Git clone operations and rotate any credentials that were present in the affected log output.\n\nReferences:\n\n- CVE-2026-110248\n- GHSA-g6w4-jvgp-68x2\n\nAcknowledgement:\n\nWe would like to thank Gouri Sankar A for collaborating on this issue through the coordinated vulnerability disclosure process.\n\n Please email aws-security@amazon.com with any security questions or concerns.","summary":"When a Git clone performed by the SageMaker Python SDK fails, the repository credentials appear in the subprocess error output and get written to CloudWatch Logs without redaction. Anyone with read access to those logs can recover the plaintext credentials.","severity":"medium","cvss":6.0,"epss":null,"exploited":false,"exploited_at":null,"products":[],"ecosystems":[],"tags":["aws","cloudwatch","lab-disclosure","package","python","sagemaker"],"not_affected_if":["fixed-version","no-credentials-in-url","no-git-clone"],"how_to_check":"- Run `pip show sagemaker` to see the installed version of the SageMaker Python SDK, and compare it with the fixed releases.\n- Search CloudWatch Logs for SageMaker jobs with Git clone failures, and look for credentials embedded in repository URLs.\n- Check whether repository URLs used by SageMaker jobs contain usernames, tokens or passwords.","fix":"- Upgrade the SageMaker Python SDK V2 to `2.257.3` or later, or V3 to `3.10.0` or later.\n- Rotate any credentials that appeared in CloudWatch Logs output from failed Git clone operations.\n- Stop putting credentials in Git repository URLs and use a secrets mechanism instead.\n- Restrict access to CloudWatch Logs that contain SageMaker SDK output.","status":"open","published_at":"2026-10-14T00:00:00+00:00","first_seen_at":"2026-10-24T20:05:36+00:00","triaged_at":"2026-10-25T04:07:54+00:00","updated_at":"2026-10-25T04:07:54+00:00","category":"package","confidence":0.8,"triage":{"at":"2026-10-25T04:07:54Z","model":"claude-haiku-5-5","output":{"fix":"- Upgrade the SageMaker Python SDK V2 to `2.257.3` or later, or V3 to `3.10.0` or later.\n- Rotate any credentials that appeared in CloudWatch Logs output from failed Git clone operations.\n- Stop putting credentials in Git repository URLs and use a secrets mechanism instead.\n- Restrict access to CloudWatch Logs that contain SageMaker SDK output.","reach":"network","title":"Amazon SageMaker Python SDK: failed Git clones log repository credentials in plaintext","impact":"medium","summary":"When a Git clone performed by the SageMaker Python SDK fails, the repository credentials appear in the subprocess error output and get written to CloudWatch Logs without redaction. Anyone with read access to those logs can recover the plaintext credentials.","category":"package","nicknames":[],"who_cares":["aws","sagemaker","python","cloudwatch"],"confidence":0.8,"needs_auth":"yes","worst_case":"Repository credentials readable by anyone who can view the CloudWatch Logs","analyst_note":"","how_to_check":"- Run `pip show sagemaker` to see the installed version of the SageMaker Python SDK, and compare it with the fixed releases.\n- Search CloudWatch Logs for SageMaker jobs with Git clone failures, and look for credentials embedded in repository URLs.\n- Check whether repository URLs used by SageMaker jobs contain usernames, tokens or passwords.","needs_analyst":false,"not_affected_if":["fixed-version","no-git-clone","no-credentials-in-url"],"who_is_affected":"Teams that run SageMaker jobs or notebooks that clone private Git repositories with the SageMaker Python SDK are exposed, most of all if credentials are embedded in the repository URL. Exposure also depends on who can read the CloudWatch Logs for those jobs. Teams on versions at or above the fixed releases, or whose SDK never performs Git clones, can skip it.","user_interaction":"no","needs_special_setup":"yes","severity_if_unknown":"medium"},"cost_usd":0.000762,"input_tokens":3969,"output_tokens":731,"severity_before":"medium"},"analyst_status":null,"analyst_note":null,"search":"'-110248':15A,18A,69C,84C,140C,287C '-134':100C '-2026':14A,17A,68C,83C,139C,286C '10/14/2026':111C '11':112C '2.257.3':196C,218C '2026':99C '3.10.0':204C,222C '30':113C '68x2':292C 'abl':181C 'access':58B,176C,256C 'acknowledg':293C 'address':210C 'affect':281C 'amazon':1A,78C,93C,117C,134C,148C,189C,197C,212C 'anyon':55B 'appear':41B 'attent':108C 'authent':172C 'avoid':247C 'aw':19B,101C,103C 'aws-security@amazon.com':315C 'bulletin':97C 'clone':7A,30B,155C,270C 'cloudwatch':20B,51B,167C,258C 'code':236C 'collabor':303C 'compon':146C 'concern':321C 'contain':260C 'content':104C 'coordin':309C 'credenti':10A,40B,66B,159C,186C,249C,275C 'cve':13A,16A,67C,82C,138C,285C 'date':110C 'deploy':129C 'deriv':235C 'descript':116C 'disclosur':23B,311C 'email':314C 'ensur':231C 'error':45B,162C 'fail':5A,37B,157C,268C 'file':76C,91C 'fix':239C 'fork':233C 'g6w4':290C 'get':48B 'ghsa':289C 'ghsa-g6w4-jvgp-68x2':288C 'git':6A,29B,154C,251C,269C 'git_utils.py':145C 'gouri':299C 'id':98C 'identifi':137C 'immedi':242C 'impact':187C 'import':106C 'includ':248C 'incorpor':237C 'inform':73C,88C 'insert':70C,85C 'issu':142C,207C,306C 'jvgp':291C 'lab':22B 'lab-disclosur':21B 'latest':228C 'learn':131C 'librari':125C 'like':296C 'log':8A,52B,61B,75C,90C,168C,179C,259C,266C,282C 'machin':130C 'model':132C 'obtain':183C 'open':123C 'oper':156C,271C 'output':46B,163C,264C,283C 'packag':24B 'pdt':115C 'perform':31B 'plaintext':12A,65B,185C 'pleas':313C 'possibl':246C 'present':278C 'process':312C 'public':109C 'python':3A,25B,35B,80C,95C,119C,150C,191C,199C,214C,262C 'question':319C 'read':57B 'recommend':224C 'recov':63B 'redact':54B,170C 'refer':284C 'remot':173C 'repositori':9A,39B,158C,252C 'requir':107C 'resolut':205C 'restrict':255C 'review':265C 'rotat':273C 'sagemak':2A,26B,34B,79C,94C,118C,135C,149C,190C,198C,213C,261C 'sankar':300C 'scope':102C 'sdk':4A,36B,81C,96C,120C,151C,192C,200C,215C,263C 'secur':318C 'sensit':72C,87C 'sourc':124C 'subprocess':44B,161C 'thank':298C 'train':127C 'type':105C 'upgrad':225C,243C 'url':253C 'user':174C 'v2':193C,216C 'v3':201C,220C 'version':188C,194C,202C,217C,221C,229C 'vulner':310C 'without':53B,169C 'workaround':240C 'would':295C 'written':49B,165C","nicknames":[],"mitigations":[],"who_affected":"Teams that run SageMaker jobs or notebooks that clone private Git repositories with the SageMaker Python SDK are exposed, most of all if credentials are embedded in the repository URL. Exposure also depends on who can read the CloudWatch Logs for those jobs. Teams on versions at or above the fixed releases, or whose SDK never performs Git clones, can skip it.","tier":"C","verified":{},"analyst_revisit_at":null,"first_public_at":"2026-10-14T00:00:00+00:00","kev":null,"cvss_vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:H/VI:N/VA:N/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","cvss_by":"ff89ba41-3aa1-4d27-914a-91399e9639e5","poc_count":0,"sources":[{"source":"lab","external_id":"https://aws.amazon.com/security/security-bulletins/rss/2026-134-aws/#CVE-2026-110248","url":"https://aws.amazon.com/security/security-bulletins/rss/2026-134-aws/","severity":null,"published_at":"2026-10-14T00:00:00+00:00","modified_at":null,"first_seen_at":"2026-10-24T20:05:36+00:00","fetched_at":"2026-10-24T20:05:36+00:00","note":"AWS: CVE-2026-110248 - Insertion of sensitive information into log files in Amazon SageMaker Python SDK"},{"source":"nvd","external_id":"CVE-2026-110248","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-110248","severity":null,"published_at":"2026-10-14T19:17:10+00:00","modified_at":"2026-10-16T15:20:39+00:00","first_seen_at":"2026-10-24T20:15:27+00:00","fetched_at":"2026-10-24T20:15:27+00:00","note":null}],"documents":[{"id":2976,"url":"https://aws.amazon.com/security/security-bulletins/rss/2026-134-aws/","title":"AWS: CVE-2026-110248 - Insertion of sensitive information into log files in Amazon SageMaker Python SDK","kind":"advisory","source":"lab","text":"CVE-2026-110248 - Insertion of sensitive information into log files in Amazon SageMaker Python SDK\n\n### CVE-2026-110248 - Insertion of sensitive information into log files in Amazon SageMaker Python SDK\n\nBulletin ID: 2026-134-AWS Scope: AWS Content Type: Important (requires attention) Publication Date: 10/14/2026 11:30 AM PDT\n\nDescription:\n\nAmazon SageMaker Python SDK is an open source library for training and deploying machine learning models on Amazon SageMaker.\n\nWe identified CVE-2026-110248, an issue in the git_utils.py component of Amazon SageMaker Python SDK. When a Git clone operation fails, repository credentials in subprocess error output are written to CloudWatch Logs without redaction. An authenticated remote user with access to those logs is able to obtain the plaintext credentials.\n\nImpacted versions:\n\n- Amazon SageMaker Python SDK V2 versions before 2.257.3\n- Amazon SageMaker Python SDK V3 versions before 3.10.0\n\nResolution:\n\nThis issue has been addressed in Amazon SageMaker Python SDK V2 version 2.257.3 and V3 version 3.10.0. We recommend upgrading to the latest version and ensuring any forked or derivative code incorporates the fixes.\n\nWorkarounds:\n\nIf immediate upgrading is not possible, avoid including credentials in Git repository URLs and restrict access to CloudWatch Logs containing SageMaker Python SDK output. Review logs from failed Git clone operations and rotate any credentials that were present in the affected log output.\n\nReferences:\n\n- CVE-2026-110248\n- GHSA-g6w4-jvgp-68x2\n\nAcknowledgement:\n\nWe would like to thank Gouri Sankar A for collaborating on this issue through the coordinated vulnerability disclosure process.\n\n Please email aws-security@amazon.com with any security questions or concerns.","published_at":"2026-10-14T00:00:00+00:00","relation":"advisory","confidence":null,"note":null}],"notice":"Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai"}