Then one of your users could install it and run it locally.

Then they could open a malicious website, e.g. something like

```
https://evilhackers.example.com
```

And that malicious website sends requests using `fetch()` with a `Blob` body to the local API at

```
http://localhost:8000/v1/agents/multivac
```

BMP, GIF, WAV, RAW, RAW\_PADDED, CSV and PROTO formats. All other input formats,
including tensorflow-io should be sandboxed if used to process untrusted data.

For example, if an attacker were to upload a malicious video file, they could
potentially exploit a vulnerability in the TensorFlow code that handles videos,
which could allow them to execute arbitrary code on the system running
TensorFlow.

    otherwise. Do not persist these and assume they can be read by a future
    version of the library.

5.  Our classes are not designed to protect against a malicious caller. You
    should not use them for communication between trusted and untrusted code.

6.  For the mainline flavor, we test the libraries using OpenJDK 8, 11, and 17