Re-structure

README.md

@@ -41,3 +41,27 @@ That is what this project proposes. We can stress-test the prompt by giving it a
 |----------------------------|-----------------------|
 | Can be used with any model | Can be time-consuming |
 | Saves possible costs       | Not 100% accurate     |
+
+## How to use
+
+Currently, this stress-test was designed to interface with [suppress.js](https://github.com/velocitatem/suppress), but it can easily be modified to work with anything else.
+
+First, you should configure the way you want to evaluate the prompts in `run.py`. In this file, you can configure the following:
+
+```python
+def runPrompt(prompt):
+  return "The evaluation of that prompt in a given model or application"
+```
+
+This function should take in a prompt and return a string that represents the evaluation of that prompt. For example, if you are using [suppress.js](https://github.com/velocitatem/suppress), you can use the existing function, and just modify the endpoint.
+
+Then, you can run the stress-test with:
+
+```bash
+python3 cps-stress-test.py
+```
+
+This will run the stress-test of the prompts on your model or application. It will output the results of the stress-test.
+
+## Contributing
+Always looking for more prompts to test! If you have any prompts that you want to add, please make a pull request. Any other contributions are also welcome.

cps-stress-test.py

@@ -2,6 +2,7 @@ import json
 import sys
 import os
 import subprocess
+from run import runPrompt
 
 # read the file malicous.csv
 # each item is like this: [malicious prompt, expected malicous response]
@@ -16,12 +17,8 @@ def readMaliciousFile():
     return malicious
 data = readMaliciousFile()
 
-import requests
+
-def runPrompt(prompt):
+
-    # send a get request to localhost:3042/new/person/:introduction
-    # the introduction is the prompt
-    r = requests.get('http://localhost:3042/capital/' + prompt)
-    return r.json().get('response')
 
 
 

run.py

@@ -0,0 +1,4 @@
+import requests
+def runPrompt(prompt):
+    r = requests.get('http://localhost:3042/capital/' + prompt)
+    return r.json().get('response')