MD5 Hash Collision in run-llama/llama_index
run-llama/llama_index
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| llama-index PyPI |
|
|
| llama-index-readers-docugami PyPI |
|
|
run-llama/llama_index
| Product | Affected | Fixed in |
|---|---|---|
| llama-index PyPI |
|
|
| llama-index-readers-docugami PyPI |
|
|
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.
From the CVE record · @huntr_ai
DocugamiReader uses MD5-based IDs that can collide when structurally distinct document chunks contain identical text, causing one chunk to overwrite another. This can lose important content, disrupt chunk hierarchies, and produce inaccurate AI responses. The input does not describe an attacker, access requirements, or user-interaction requirements.
Check installations of PyPI llama-index below 0.12.41 or llama-index-readers-docugami below 0.3.1 that use DocugamiReader.