Initial clean commit without secrets

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+.env

README.md

@@ -0,0 +1,13 @@
+---
+title: Hackathon
+emoji: 🏆
+colorFrom: indigo
+colorTo: green
+sdk: gradio
+sdk_version: 5.34.2
+app_file: app.py
+pinned: false
+short_description: hf space for generating legal analysis for hack the law
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,218 @@
+from huggingface_hub import login
+from smolagents import CodeAgent, DuckDuckGoSearchTool, HfApiModel, tool, Tool, load_tool, InferenceClientModel
+from smolagents.models import ChatMessage
+from transformers import pipeline
+import cohere
+from gradio_client import Client
+from newsapi import NewsApiClient
+import requests
+import gradio as gr
+import os
+from dotenv import load_dotenv
+import json
+from transformers import pipeline
+
+load_dotenv()
+
+newsApiKey = os.getenv('NEWSAPI_KEY')
+#grok_api_key = os.getenv('GROK_API_KEY')
+HF_TOKEN = os.getenv("HF_TOKEN")
+login(token=HF_TOKEN)
+COHERE_API_KEY = os.getenv('COHERE_API_KEY')
+
+from groq import Groq
+
+client = Groq(
+    api_key=os.environ.get("GROQ_API_KEY"),
+)
+
+
+@tool
+def get_company_news_articles(company_name: str) -> dict:
+    """
+    fetch news articles about a company.
+
+    Args:
+        company_name (str): the company to get news articles about
+    
+    Returns:
+        list: a list of articles about the company
+    """
+
+    response = requests.get(f'https://newsapi.org/v2/everything?q={company_name}&from=2025-06-01&sortBy=popularity&apiKey={newsApiKey}')
+    try:
+        data = response.json()
+    except Exception as e:
+        print("❌ Failed to parse JSON:", str(e))
+        return []
+
+    print("🔍 Full NewsAPI response:", data)
+
+    if data.get("status") != "ok":
+        print("❌ NewsAPI error:", data.get("message"))
+        return []
+
+
+    print(f"results from news api get request are: {data.get('articles')}")
+
+    articles = data.get('articles', [])
+    filtered_articles = [article for article in articles]
+    return filtered_articles
+    
+
+
+class OllamaLLM:
+    pass
+    #just borrow all code below, and switched out the api call for the api responses to localhost port
+
+class grokLLM:
+    pass
+
+class mistralLLM:
+    pass
+
+
+
+class CohereLLM:
+    def __init__(self, api_key: str = COHERE_API_KEY, model_name: str = "command-r-plus"):
+        self.client = cohere.Client(api_key)
+        self.model_name = model_name
+
+    def __call__(self, prompt: str, **kwargs) -> str:
+        # Remove keys not supported by Cohere
+        kwargs.pop('stop_sequences', None)
+
+        if isinstance(prompt, list):
+            prompt = self._convert_chat_to_prompt(prompt)
+
+        # Optional: set temperature, max tokens, etc.
+        temperature = kwargs.get("temperature", 0.7)
+        max_tokens = kwargs.get("max_tokens", 300)
+
+        response = self.client.generate(
+            model=self.model_name,
+            prompt=prompt,
+            max_tokens=max_tokens,
+            temperature=temperature,
+        )
+        #return response.generations[0].text.strip()
+
+        output_text = response.generations[0].text.strip()
+
+        # ✅ Wrap response in ChatMessage
+        return ChatMessage(role="assistant", content=output_text)
+    
+    def _convert_chat_to_prompt(self, messages):
+        """
+        Convert chat-style message history to a single string prompt.
+        Example input: [{"role": "user", "content": "Hello"}, {"role": "assistant", "content": "Hi!"}]
+        """
+        prompt = ""
+        for message in messages:
+            role = message.get("role", "user")
+            content = message.get("content", "")
+            if role == "user":
+                prompt += f"User: {content}\n"
+            elif role == "assistant":
+                prompt += f"Assistant: {content}\n"
+            else:
+                prompt += f"{role.title()}: {content}\n"
+        prompt += "Assistant:"  # Final cue for model to respond
+        return prompt
+
+
+
+
+cohere_model = CohereLLM()
+
+
+
+
+
+#HfApiModel("mistralai/Mistral-7B-v0.1-chat")
+
+def mainFunc(articles):
+    articles = json.loads(articles)
+    newsApiKey = os.getenv('NEWSAPI_KEY')
+    if not newsApiKey:
+        raise ValueError("Missing NEWS_API_KEY in environment variables.")
+    
+    print(f"data structure of articles is: {articles}")
+
+    prompt = f"""
+You are an agent that analyzes the risk factors for a company, by using the data from:
+- Documents: {articles['company_info']['documents']}
+- News summaries: {articles['news_data']['articles_summary']}
+
+Please respond using **only** the following JSON format:
+
+{{
+  "risk_factors": [
+    {{
+      "severity": "Low | Medium | High",
+      "specific_event": "the specific type of event, i. e. nuclear war"
+      "risk_type": "e.g., Operational Risk, Financial Risk, Reputational Risk, Legal Risk",
+      "affected_contracts": "Name or path of the affected document",
+      "affected_clauses": "Clause number(s) or section(s)",
+      "narrative": {{
+        "solutions_in_contract": "Summarize contractual protections or remedies",
+        "alternative_mitigation_strategies": "Suggest other ways to reduce the risk",
+        "monitoring_tasks": "Define ongoing monitoring or reporting actions"
+      }}
+    }}
+  ],
+  "available_data": {{
+    "context_items": ["Summary of the company info and risk-related context"],
+    "document_count": {len(articles['company_info']['documents'])},
+    "content_available": true,
+    "news_available": true
+  }}
+}}
+"""
+
+
+    chat_completion = client.chat.completions.create(
+    messages=[
+        {
+            "role": "system",
+            "content": prompt
+        },
+        {
+            "role": "user",
+            "content": f"What are the {articles['analysis_request']['analysis_scope']} risk factors for {articles['company_info']['name']} ?",
+        }
+    ],
+    model="llama-3.3-70b-versatile",
+    response_format={"type": "json_object"},#and include word 'json' in messages/prompt
+    )
+    print(chat_completion.choices[0].message.content)
+    return chat_completion.choices[0].message.content
+
+    
+    
+    return result
+
+
+#add agents it can hand off to
+
+
+
+
+
+#agent.prompt_templates["system_prompt"] = agent.prompt_templates["system_prompt"] + "\n when asked for most recent articles, return each article with its dict/list values, rather than just the title"
+#agent.run("what are the most recent articles about Microsoft?")
+#print(agent.prompt_templates["system_prompt"])
+
+
+#huggingface-cli login - to set access token in temrainl and save it
+#translation function works well
+
+demo = gr.Interface(
+    fn=mainFunc,
+    inputs="text",
+    outputs="text",
+    title="hackathon agent",
+    description="finds info about a company"
+)
+
+demo.launch(share=True)