Remote Code Execution via Prompt Injection in vanna-ai/vanna
Key facts
- Attack vector
- Network
- Privileges required
- None
- User interaction
- None
- Attack complexity
- Low
- Impact type
- Code execution
- Confidentiality · integrity · availability
- High · High · High
- Fix status
- No fix listed
Attack profile from CVSS 3.0 · CNA. Fix status from CNA and GitHub. Impact type from CWE-94.
Summary
In the latest version of vanna-ai/vanna, the `vanna.ask` function is vulnerable to remote code execution due to prompt injection. The root cause is the lack of a sandbox when executing LLM-generated code, allowing an attacker to manipulate the code executed by the `exec` function in `src/vanna/base/base.py`. This vulnerability can be exploited by an attacker to achieve remote code execution on the app backend server, potentially gaining full control of the server.
Why it matters now
SSVC decision points
CISA ADP · SSVC 2.0.3
- Exploitation
- Poc
- Automatable
- No
- Technical impact
- Total
The decision (Track, Attend, Act) also depends on your mission and well-being impact; see CISA's SSVC guide.
Affected products
| Vendor | Product | Affected versions | Fixed in | Status | Source |
|---|---|---|---|---|---|
| vanna-ai | vanna-ai/vanna | up to latest | — | No fix listed | CNA |
| — | vanna (PyPI) | up to 0.6.2 | — | No fix listed | GHSA-rrqq-fv6m-692m |
- vanna-ai/vanna Vendorvanna-ai Affected up to latest Fixed in — No fix listed CNA
- vanna (PyPI) Vendor— Affected up to 0.6.2 Fixed in — No fix listed GHSA-rrqq-fv6m-692m
CPE names CISA-ADP · 1
| CPE | Versions | Role | Vendor |
|---|---|---|---|
| cpe:2.3:a:vanna-ai:vanna:-:*:*:*:*:*:*:* | not specified by CISA-ADP | — | — |
- cpe:2.3:a:vanna-ai:vanna:-:*:*:*:*:*:*:* Versions not specified by CISA-ADP — ·—
Remediation
No fixed version listed by the sources yet.
Timeline
- 2026-10-02 Threatist record created from CVE-2024-5826
- 2026-10-03 Alias added: GHSA-rrqq-fv6m-692m
References
More vanna-ai vulnerabilities
- TI-2024-05403 Arbitrary File Write by Prompt Injection via DuckDB SQL in vanna-ai/vanna CVSS 3.0CRITICAL 9.8
- TI-2026-46836 Vanna 2.0.2 Path Traversal via FileSystemConversationStore CVSS 4.0HIGH 8.8
- TI-2024-07510 Server-Side Request Forgery (SSRF) in vanna-ai/vanna CVSS 3.0HIGH 8.3
- TI-2024-07469 Local File Read (LFI) by Prompt Injection via SnowFlake SQL in vanna-ai/vanna CVSS 3.0HIGH 7.5
- TI-2024-05342 Local File Read (LFI) by Prompt Injection via Postgres SQL in vanna-ai/vanna CVSS 3.0HIGH 7.5
- TI-2026-06105 vanna-ai vanna Legacy Flask API improper authorization CVSS 4.0MEDIUM 6.9
Sources
- CVE-2024-5826 · CVE Program
- CVE-2024-5826 · NVD
- FIRST EPSS
- GHSA-rrqq-fv6m-692m · GitHub Advisory Database · CC BY 4.0
This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences