Spaces:
Sleeping
Sleeping
Commit ·
c296117
1
Parent(s): 6717b72
feat : renamed everything to breach OS
Browse files- docs/reward_design.md +1 -1
- explaination.md +2 -3
- server/config.py +1 -1
docs/reward_design.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
# Reward Design
|
| 2 |
|
| 3 |
-
This document details the reward functions and grading logic for the OpenEnv
|
| 4 |
|
| 5 |
## 1. Difficulties and Graders
|
| 6 |
|
|
|
|
| 1 |
# Reward Design
|
| 2 |
|
| 3 |
+
This document details the reward functions and grading logic for the OpenEnv BreachOS environment.
|
| 4 |
|
| 5 |
## 1. Difficulties and Graders
|
| 6 |
|
explaination.md
CHANGED
|
@@ -1,7 +1,6 @@
|
|
| 1 |
-
#
|
| 2 |
|
| 3 |
-
|
| 4 |
-
**RedTeamOS** is an AI Red-Teaming Environment for Safety Research. The project is structured as a client-server architecture built in Python, where an automated attacker agent repeatedly tries to bypass a conversational defender LLM across multi-turn episodes.
|
| 5 |
|
| 6 |
The architecture is divided among a 3-person team:
|
| 7 |
1. **Person 1 (The Architect)**: Built the core structural pieces, Pydantic data schemas, FastAPI server, Dockerization, and the automated HuggingFace test scripts.
|
|
|
|
| 1 |
+
# BreachOS Codebase Explanation
|
| 2 |
|
| 3 |
+
**BreachOS** is an AI Red-Teaming Environment for Safety Research. The project is structured as a client-server architecture built in Python, where an automated attacker agent repeatedly tries to bypass a conversational defender LLM across multi-turn episodes.
|
|
|
|
| 4 |
|
| 5 |
The architecture is divided among a 3-person team:
|
| 6 |
1. **Person 1 (The Architect)**: Built the core structural pieces, Pydantic data schemas, FastAPI server, Dockerization, and the automated HuggingFace test scripts.
|
server/config.py
CHANGED
|
@@ -2,7 +2,7 @@ from pydantic_settings import BaseSettings
|
|
| 2 |
from functools import lru_cache
|
| 3 |
|
| 4 |
class Settings(BaseSettings):
|
| 5 |
-
app_name: str = "
|
| 6 |
debug: bool = False
|
| 7 |
max_turns: int = 10
|
| 8 |
|
|
|
|
| 2 |
from functools import lru_cache
|
| 3 |
|
| 4 |
class Settings(BaseSettings):
|
| 5 |
+
app_name: str = "BreachOS"
|
| 6 |
debug: bool = False
|
| 7 |
max_turns: int = 10
|
| 8 |
|