Description
Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by appending control operators like semicolons or logical operators to approved commands, bypassing the session approval requirement.
Affected products
| Vendor | Product | Affected | Fixed |
|---|---|---|---|
| ollama | ollama | 0.14.0 – < 0.31.2 | 0.31.2 |
Public exploits
Exploit or proof-of-concept code for this vulnerability is publicly available, which lowers the bar for attackers. Treat it as a reason to patch sooner.
GitHub 1 repository
-
akinerkisa/blindllama-cve-2026-102697 ★ 0 · Oct 15, 2026
BlindLlama: CVE-2026-102697 Reproduction
Sources: PoC-in-GitHub, Metasploit Framework, Exploit-DB — updated daily. Only links are listed; no exploit code is hosted here.
Advisories 1
- WID-SEC-2026-3634 Ollama: Schwachstelle ermöglicht Codeausführung BSI CERT-Bund (WID) · Sep 29, 2026
References
- Patch Commit patch
- ollama v0.31.2 Release Notes release-notes
- https://github.com/ollama/ollama/blob/v0.31.1/x/agent/approval.go#L204-L206 technical-description
- https://github.com/ollama/ollama/blob/v0.31.1/x/agent/approval.go#L389 technical-description
- https://github.com/ollama/ollama product
- VulnCheck Advisory: Ollama 0.14.0 before 0.31.2 Experimental Agent Bash Approval Bypass via Prefix-Based Authorization third-party-advisory