# LlamaIndex reports AI-agent security tests of LlamaParse over four days

> Original title: We gave AI agents a standard login and asked them to break LlamaParse.

- Source: LlamaIndex
- Published: 2026-11-11T17:52:49.000Z
- Bad Influence: https://badinfluence.news/items/xf4nuik4v4ekwtri2ayw3jip9
- Original: https://x.com/llama_index/status/2120575489752568097

## Summary

LlamaIndex says it gave AI agents a standard login and asked them to break LlamaParse over four days. The agents tested parsing, extraction, indexing, deployed agents and its MCP server. The company reports that model behavior varies between runs, so a single test pass is not enough, and that extraction reads text a person cannot see on the page.

## Post

We gave AI agents a standard login and asked them to break LlamaParse.

Over four days, they tested every step a document goes through: parsing, extraction, indexing, deployed agents and our MCP server.

What we learned:
1/ Model behavior varies between runs, so a single test pass is not enough
2/ Some vulnerabilities only appear once you build a real application
3/ Defining expected behavior up front separates real issues from noise
4/ Extraction reads everything on the page, including text a person can’t see

Our engineering team wrote up the full breakdown below.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHW3MKXQawAACnaK.jpg&mode=thumb&exp=1794700800&sig=15f0f16fc6fded54)