> For the complete documentation index, see [llms.txt](https://docs.perkinsfund.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.perkinsfund.org/readme/pcef-public-endpoints/aura.md).

# AURA API

AURA is a lightweight malware triage API that classifies uploaded executable files and lets you look up recent or specific scan results by SHA-256. Currently supports Windows PE and Linux ELF; other formats return `unknown`.

* [File Prediction](#file-prediction-endpoint)
* [Latest Scans](#latest-scans-endpoint)
* [Search Scans](#search-scans-endpoint)

### API BASE URL: `https://ai.perkinsfund.org`

***

### File Prediction Endpoint

Endpoint: `/api/predict`

Request type: POST

Data type: File

Headers (optional): X-Api-Key: API KEY

**NOTE: The AI currently only classifies ELF and Windows PE executable file formats. Others will be "unknown".**

#### Example request

```bash
curl -F"filename=@/path/to/file" https://ai.perkinsfund.org/api/predict
```

#### Expected outputs

Success:

```json
{
  ...,
    "results": {
    "class": "safe",
    "created_at": 1743103981.5979574,
    "file_hash": "ed369187681a62247e38d930320f1cd771756d0b7b67072d8ec655ef99e14aeb",
    "time_taken": 0.37021786579862237
  },
  ...
}
```

Unsuccessful:

```json
{
  ...,
  {
   "error": {
      "error_message": "The message"
   }, 
   ...
}
```

***

### Latest Scans Endpoint

Endpoint: `/api/latest`

Request type: POST

Data type:

Headers (optional): X-Api-Key: API KEY

**NOTE: This POST request does not take any parameters, it will return the last 50 scans**

#### Example request

```bash
curl -X POST https://ai.perkinsfund.org/api/latest
```

#### Expected outputs

Success:

```json
{
  ...,
  [
    {
      "class": "unknown",
      "created_at": 1743103290.500192,
      "error": {},
      "file_hash": "008632ed4a903dbc95d7a0d042d77df9bf651f10a26a8d4557eb9d2533193ad1",
      "time_taken": 0.0004930980503559113
    },
    {
      "class": "unknown",
      "created_at": 1743099196.4861066,
      "error": {},
      "file_hash": "f5f5ac913edc3fa09bc71ff99d6b84fa63c5987f9713fdb1d55f590ac283b263",
      "time_taken": 0.000505556003190577
    }
  ],
  ...
}
```

Unsuccessful:

```json
{
  ...,
  {
    "error_message": {"error_message": "..."},
    "results": []
  },
  ...
}
```

***

### Search Scans Endpoint

Endpoint: `/api/search`

Request type: POST

Data type: JSON

Headers (required): X-Api-Key: API KEY

**NOTE: This endpoint requires an API key. You can get an API key here: <https://perkinsfund.org/login>**

#### Example request

```bash
curl -XPOST -H "x-api-key: USER KEY" \
  -H "content-type: application/json" \
  --data '{"sha256":"USER SHA256 HASH"}' \
  https://ai.perkinsfund.org/api/search
```

#### Expected outputs

Success:

```json
{
  ...,
  "results": {
    "class": "safe",
    "created_at": 1742832170.874544,
    "error": {},
    "file_hash": "975d2ab9067a1b21a46ce9d87e6ab589636d128fcba4f49e53392815ebc72d77",
    "time_taken": 0.7284463751129806
  },
  ...
}
```

Unsuccessful:

```json
{
  ...,
  "error": {
    "error_message": "SHA256 hash did not pass heuristics check, is it a hash?" OR "invalid API key supplied"
  },
  ...
}
```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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, and the optional `goal` query parameter:

```
GET https://docs.perkinsfund.org/readme/pcef-public-endpoints/aura.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.