PyPI package
llama-index-core
pkg:pypi/llama-index-core
Vulnerabilities (10)
| CVE | Sev | CVSS | KEV | Affected versions | Fixed in | Published | Description |
|---|---|---|---|---|---|---|---|
| CVE-2025-6208 | Med | 5.3 | < 0.12.41 | 0.12.41 | Feb 2, 2026 | The `SimpleDirectoryReader` component in `llama_index.core` version 0.12.23 suffers from uncontrolled memory consumption due to a resource management flaw. The vulnerability arises because the user-specified file limit (`num_files_limit`) is applied after all files in a directory | |
| CVE-2025-7647 | Hig | 7.3 | < 0.13.0 | 0.13.0 | Sep 27, 2025 | The llama-index-core package, up to version 0.12.44, contains a vulnerability in the `get_cache_dir()` function where a predictable, hardcoded directory path `/tmp/llama_index` is used on Linux systems without proper security controls. This vulnerability allows attackers on multi | |
| CVE-2025-5302 | Hig | 8.6 | < 0.12.38 | 0.12.38 | Aug 25, 2025 | A denial of service vulnerability exists in the JSONReader component of the run-llama/llama_index repository, specifically in version v0.12.37. The vulnerability is caused by uncontrolled recursion when parsing deeply nested JSON files, which can lead to Python hitting its maximu | |
| CVE-2025-6209 | — | >= 0.11.23, < 0.12.41 | 0.12.41 | Jul 7, 2025 | A path traversal vulnerability exists in run-llama/llama_index versions 0.12.27 through 0.12.40, specifically within the `encode_image` function in `generic_utils.py`. This vulnerability allows an attacker to manipulate the `image_path` input to read arbitrary files on the server | ||
| CVE-2025-5472 | — | < 0.12.38 | 0.12.38 | Jul 7, 2025 | The JSONReader in run-llama/llama_index versions 0.12.28 is vulnerable to a stack overflow due to uncontrolled recursive JSON parsing. This vulnerability allows attackers to trigger a Denial of Service (DoS) by submitting deeply nested JSON structures, leading to a RecursionError | ||
| CVE-2025-3108 | — | >= 0.11.15, < 0.12.41 | 0.12.41 | Jul 6, 2025 | A critical deserialization vulnerability 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. JsonPickleSe | ||
| CVE-2024-12704 | — | < 0.12.6 | 0.12.6 | Mar 20, 2025 | A vulnerability in the LangChainLLM class of the run-llama/llama_index repository, version v0.12.5, allows for a Denial of Service (DoS) attack. The stream_complete method executes the llm using a thread and retrieves the result via the get_response_gen method of the StreamingGen | ||
| CVE-2024-45201 | — | < 0.10.38 | 0.10.38 | Aug 22, 2024 | An issue was discovered in llama_index before 0.10.38. download/integration.py includes an exec call for import {cls_name}. | ||
| CVE-2024-3271 | — | < 0.10.24 | 0.10.24 | Apr 16, 2024 | A command injection vulnerability exists in the run-llama/llama_index repository, specifically within the safe_eval function. Attackers can bypass the intended security mechanism, which checks for the presence of underscores in code generated by LLM, to execute arbitrary code. Th | ||
| CVE-2024-3098 | Cri | 9.8 | < 0.10.24 | 0.10.24 | Apr 10, 2024 | A vulnerability was identified in the `exec_utils` class of the `llama_index` package, specifically within the `safe_eval` function, allowing for prompt injection leading to arbitrary code execution. This issue arises due to insufficient validation of input, which can be exploite |
- affected < 0.12.41fixed 0.12.41
The `SimpleDirectoryReader` component in `llama_index.core` version 0.12.23 suffers from uncontrolled memory consumption due to a resource management flaw. The vulnerability arises because the user-specified file limit (`num_files_limit`) is applied after all files in a directory
- affected < 0.13.0fixed 0.13.0
The llama-index-core package, up to version 0.12.44, contains a vulnerability in the `get_cache_dir()` function where a predictable, hardcoded directory path `/tmp/llama_index` is used on Linux systems without proper security controls. This vulnerability allows attackers on multi
- affected < 0.12.38fixed 0.12.38
A denial of service vulnerability exists in the JSONReader component of the run-llama/llama_index repository, specifically in version v0.12.37. The vulnerability is caused by uncontrolled recursion when parsing deeply nested JSON files, which can lead to Python hitting its maximu
- CVE-2025-6209Jul 7, 2025affected >= 0.11.23, < 0.12.41fixed 0.12.41
A path traversal vulnerability exists in run-llama/llama_index versions 0.12.27 through 0.12.40, specifically within the `encode_image` function in `generic_utils.py`. This vulnerability allows an attacker to manipulate the `image_path` input to read arbitrary files on the server
- CVE-2025-5472Jul 7, 2025affected < 0.12.38fixed 0.12.38
The JSONReader in run-llama/llama_index versions 0.12.28 is vulnerable to a stack overflow due to uncontrolled recursive JSON parsing. This vulnerability allows attackers to trigger a Denial of Service (DoS) by submitting deeply nested JSON structures, leading to a RecursionError
- CVE-2025-3108Jul 6, 2025affected >= 0.11.15, < 0.12.41fixed 0.12.41
A critical deserialization vulnerability 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. JsonPickleSe
- CVE-2024-12704Mar 20, 2025affected < 0.12.6fixed 0.12.6
A vulnerability in the LangChainLLM class of the run-llama/llama_index repository, version v0.12.5, allows for a Denial of Service (DoS) attack. The stream_complete method executes the llm using a thread and retrieves the result via the get_response_gen method of the StreamingGen
- CVE-2024-45201Aug 22, 2024affected < 0.10.38fixed 0.10.38
An issue was discovered in llama_index before 0.10.38. download/integration.py includes an exec call for import {cls_name}.
- CVE-2024-3271Apr 16, 2024affected < 0.10.24fixed 0.10.24
A command injection vulnerability exists in the run-llama/llama_index repository, specifically within the safe_eval function. Attackers can bypass the intended security mechanism, which checks for the presence of underscores in code generated by LLM, to execute arbitrary code. Th
- affected < 0.10.24fixed 0.10.24
A vulnerability was identified in the `exec_utils` class of the `llama_index` package, specifically within the `safe_eval` function, allowing for prompt injection leading to arbitrary code execution. This issue arises due to insufficient validation of input, which can be exploite