AI AgentHigh
CVE-2026-82233: SiYuan before v3.8.1 contains a path traversal vulnerability in the asset.upload MCP tool that accep…
- Occurred
- Aug 28, 2026
- Published
- Aug 28, 2026
- Source
- NVD CVE
Summary
SiYuan before v3.8.1 contains a path traversal vulnerability in the asset.upload MCP tool that accepts arbitrary absolute file paths without workspace boundary validation. Attackers can induce the AI Agent to upload sensitive files such as SSH keys or credentials from outside the workspace into the asset directory through prompt injection.
Tags
- CVE
- CVE-2026-82233
- cvss:MEDIUM
- agent-fail:injection-direct
- agent-fail:authorization
- agent-cause:prompt-design
- agent-cause:missing-guardrail
Classification
Category confidence: 95%
Severity confidence: 90%
Rationale: AI agent tool vulnerability allows path traversal and unauthorized file access via prompt injection.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8