CVE-2025-1750: LlamaIndex DuckDBVectorStore vulnerable to SQL injection through ref_doc_id
CVE-2025-1750 · Severity: critical · CVSS 9.8 · Published 2025-06-02
An SQL injection vulnerability exists in the delete function of DuckDBVectorStore in run-llama/llama_index version v0.12.19. This vulnerability allows an attacker to manipulate the ref_doc_id parameter, enabling them to read and write arbitrary files on the server, potentially leading to remote code execution (RCE).
Yes. llama-index (PyPI): upgrade to 0.12.21 or later; llama-index-vector-stores-duckdb (PyPI): upgrade to 0.12.21 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
Am I affected by CVE-2025-1750?
Check whether you run llama-index (PyPI) or llama-index-vector-stores-duckdb (PyPI) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.
How to fix
llama-index (PyPI): upgrade to 0.12.21 or later
pip install "llama-index>=0.12.21"
llama-index-vector-stores-duckdb (PyPI): upgrade to 0.12.21 or later