Vulnerability · Published Jul 10, 2025 · Assigned by @huntr_ai
CVE-2025-6211MD5 Hash Collision in run-llama/llama_index
Medium 6.5
In plain English
A medium-severity security flaw in run-llama/llama_index from run-llama. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.4% (higher than 30% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.
Official description
A vulnerability in the DocugamiReader class of the run-llama/llama_index repository, up to version 0.12.28, involves the use of MD5 hashing to generate IDs for document chunks. This approach leads to hash collisions when structurally distinct chunks contain identical text, resulting in one chunk overwriting another. This can cause loss of semantically or legally important document content, breakage of parent-child chunk hierarchies, and inaccurate or hallucinated responses in AI outputs. The issue is resolved in version 0.3.1.
Affected packages & fixed versions
| Ecosystem | Package | Fixed in | Advisory |
|---|---|---|---|
| PyPI | llama-index | 0.12.41 | GHSA-5hq9-5r78-2gjh |
| PyPI | llama-index | 0.12.41 | PYSEC-2026-2606 |
| PyPI | llama-index-readers-docugami | 0.3.1 | GHSA-5hq9-5r78-2gjh |
| PyPI | llama-index-readers-docugami | 0.3.1 | PYSEC-2026-1566 |
Affected products (as reported)
| Vendor | Product | Versions |
|---|---|---|
| run-llama | run-llama/llama_index | unspecified – before 0.3.1 |
Weakness: CWE-440 · Vector: CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:L