OpenHuman: Yuanbao non-web channel can bypass human approval to execute `node_exec`

Özet

Türkçe özeti henüz hazır değil; şimdilik kaynağın kendi açıklamasını gösteriyoruz.

OpenHuman is an open-source agent harness available as a desktop application and Rust library. Before 0.57.13, a remote direct message received through a configured Yuanbao channel is converted into a generic channel AgentTurnRequest that uses the default tool policy without establishing an interactive approval context. Under the default Supervised autonomy policy with Node enabled, node_exec is classified as a write operation that should prompt for approval, but ApprovalGate treats missing APPROVAL_CHAT_CONTEXT as a non-interactive turn and allows execution. Because Yuanbao direct messages are open by default, an external sender who can message the connected bot can influence the agent to execute JavaScript and operating-system commands as the OpenHuman user without an approval prompt. The issue requires a configured Yuanbao channel, the default approval-requiring policy state, Node availability, and successful model selection of node_exec. This issue is fixed in version 0.57.13.

Etkilenen sürümler1 yazılım

Kaynaklar6 kaynak

• CVE.org kaydı
• NVD kaydı
• github.com/…nhuman/security/advisories/GHSA-gg3m-wgmq-vhmx Diğer
• github.com/tinyhumansai/openhuman/pull/3227 Diğer
• github.com/…ommit/3a7180da7765af861bf44d79e025fc67dbeae0ca Diğer
• github.com/tinyhumansai/openhuman/releases/tag/v0.57.13 Diğer

Kayıt geçmişi

1. 5 Kas 2026 19:35Yayımlandı

İlgili açıklar

• 7,7 CVE-2026-49407OpenHuman: Unauthenticated dictation WebSocket allows arbitrary audio streaming and unbounded memory growth
• 8,7 CVE-2026-54142OpenHuman: Indirect prompt injection via background email sync executes Composio actions without user approval
• 8,8 CVE-2026-54146OpenHuman: Bearer token leak via process environment variable on macOS (same-UID local RCE primitive)
• 9,8 CVE-2026-54143OpenHuman allows remote inbound channel messages to create unapproved shell cron jobs leading to host command execution
• 8,4 CVE-2026-49393OpenHuman: Unauthenticated Socket.IO RPC bypass allows invoking protected OpenHuman core controllers without the Bearer token
• 8,8 CVE-2026-54145OpenHuman: Bearer token leaked via approval gate session_id (SQLite plaintext + debug log + RPC response)