# Lineaje AI TAR (Threat Analysis Report)

Welcome to Lineaje AI TAR (Threat Analysis Report) where we focus on risks, abuse patterns, and exploitation techniques targeting AI systems, including large language models, MCP servers, and AI agents. Each advisory documents observed or plausible threats, maps them to an AI-specific kill chain, and provides guidance for detection, impact assessment, and mitigation.

Our goal is to help security, engineering, and governance teams understand how AI incidents emerge, evolve, and can be prevented—before they turn into production failures or policy violations.

## Recent Threats

<table data-view="cards"><thead><tr><th></th><th></th><th></th><th data-hidden data-card-cover data-type="image">Cover image</th><th data-hidden></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><h4><i class="fa-diamond-exclamation">:diamond-exclamation:</i></h4></td><td>Reasoning Hijack</td><td>$5M in fraudulent transactions!</td><td><a href="https://1488616604-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FeuEBKCXHKwZVjAGc0tuF%2Fuploads%2FkqCDFdcM4XXzMWe3ZTkr%2FReasoning%20Hijack%20Hero%201.png?alt=media&#x26;token=940bf9a8-239b-4cb8-8a72-dd7ddeaf4b9e">Reasoning Hijack Hero 1.png</a></td><td></td><td><a href="reasoning-and-goal-integrity/reasoning-hijack">reasoning-hijack</a></td></tr><tr><td><h4><i class="fa-diamond-exclamation">:diamond-exclamation:</i></h4></td><td>Reprompting</td><td>The single-click Microsoft Copilot attack</td><td><a href="https://1488616604-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FeuEBKCXHKwZVjAGc0tuF%2Fuploads%2FtTMXM17oroQnJZZt5Dy1%2Fimage%20(1).jpg?alt=media&#x26;token=eac089f2-0b89-4adb-97aa-2e6de6c6dc62">image (1).jpg</a></td><td></td><td><a href="input-and-instruction-manipulation/reprompt">reprompt</a></td></tr><tr><td><h4><i class="fa-diamond-exclamation">:diamond-exclamation:</i></h4></td><td>Remote Code Execution</td><td>DockerDash AI‑Driven Metadata Abuse</td><td><a href="https://1488616604-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FeuEBKCXHKwZVjAGc0tuF%2Fuploads%2FwwuMplXmcxoFPSrvHU8Q%2FDocker%20RCE%203.png?alt=media&#x26;token=0094ac4d-2757-4fd4-9815-2d59a3768f3f">Docker RCE 3.png</a></td><td></td><td><a href="instruction-manipulation/the-obedient-accomplice">the-obedient-accomplice</a></td></tr></tbody></table>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.veedna.com/lineaje-ai-threat-advisory/readme.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.