CVE ID :CVE-2026-48797<\/p>\n
Published : June 16, 2026, 11:35 p.m. | 2\u00a0hours, 7\u00a0minutes ago<\/p>\n
Description :Backpropagate is a Python library for fine-tuning large language models on a single GPU. In versions 1.1.0 and 1.1.1, the optional Reflex web UI exposes a training control plane without authentication: dataset upload, model load, training start\/stop, multi-run orchestration, GGUF export, and HuggingFace Hub push. The CLI accepts two operator-facing flags intended as security controls: –auth user:pass \u2014 documented as “require HTTP Basic authentication on every request to the UI.” and–share \u2014 documented as “expose the UI on a public address; requires –auth.” When –auth user:pass is passed, the CLI prints Auth: enabled (user: ) to confirm to the operator that authentication is active, then exports BACKPROPAGATE_UI_AUTH=user:pass to the subprocess that launches the Reflex backend. The Reflex backend (backpropagate\/ui_app\/**) never reads BACKPROPAGATE_UI_AUTH. No authentication middleware is registered. No request-level guard runs. No WebSocket upgrade guard runs. Any client that reaches the bound port \u2014 local or remote, depending on whether –share is used \u2014 has full UI access. An inline comment at backpropagate\/cli.py:1217-1218 in the v1.1.0 source documents the gap: “For Phase 1 the variable is exported but Reflex doesn’t read it yet.” This comment was internal-facing; the user-facing documentation (README, CHANGELOG, SHIP_GATE) advertised the contract as enforced. An attacker who reaches the bound port can read uploaded datasets, trigger arbitrary training runs against any local base models as well as read their paths, trigger HuggingFace Hub pushes and cause disk-fill DoS. This issue has been fixed in version 1.2.0.<\/p>\n
Severity: 0.0 | NA<\/p>\n
Visit the link for more details, such as CVSS details, affected products, timeline, and more…\u00a0<\/p>\n","protected":false},"excerpt":{"rendered":"
CVE ID :CVE-2026-48797 Published : June 16, 2026, 11:35 p.m. | 2\u00a0hours, 7\u00a0minutes ago Description :Backpropagate is a Python library for fine-tuning large language models on a single GPU. In versions 1.1.0 and 1.1.1, the optional Reflex web UI exposes a training control plane without authentication: dataset upload, model load, training start\/stop, multi-run orchestration, GGUF …<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[26],"tags":[],"class_list":["post-80661","post","type-post","status-publish","format-standard","hentry","category-vulnerability"],"_links":{"self":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/posts\/80661","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/comments?post=80661"}],"version-history":[{"count":0,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/posts\/80661\/revisions"}],"wp:attachment":[{"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/media?parent=80661"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/categories?post=80661"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/afaghhosting.net\/blog\/wp-json\/wp\/v2\/tags?post=80661"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}