Free AI Safety Workshop — in collaboration with explainx.aiSave your seat
AgentbeamBeta
AI agent vulnerability · LlamaIndex

CVE-2025-3108: LlamaIndex has Incomplete Documentation of Program Execution related to JsonPickleSerializer component

mediumCVSS 3.0 5.0GHSA-m84c-4c34-28gf

Incomplete Documentation of Program Execution exists in the run-llama/llamaindex library's JsonPickleSerializer component, affecting versions v0.12.27 through v0.12.40. This vulnerability allows remote code execution due to an insecure fallback to Python's pickle module. JsonPickleSerializer prioritizes deserialization using pickle.loads(), which can execute arbitrary code when processing untrusted data.

Published Jul 7, 2025

Affected packages

llama-index-core

Ecosystem
PyPI
Affected
>= 0.11.15, <= 0.12.40
Fixed in
0.12.41
Check
pip show llama-index-core

Advisory details

From the GitHub Advisory Database (CC BY 4.0).

Incomplete Documentation of Program Execution exists in the run-llama/llama_index library's JsonPickleSerializer component, affecting versions v0.12.27 through v0.12.40. This vulnerability allows remote code execution due to an insecure fallback to Python's pickle module. JsonPickleSerializer prioritizes deserialization using pickle.loads(), which can execute arbitrary code when processing untrusted data. Attackers can exploit this by crafting malicious payloads to achieve full system compromise. The root cause involves the use of an insecure fallback strategy without sufficient input validation or protective safeguards. Version 0.12.41 renames JsonPickleSerializer to PickleSerializer and adds a warning to the docs to only use PickleSerializer to deserialize safe things.

References

How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology

More LlamaIndex vulnerabilities

All LlamaIndex entries