From 4480ead6aaa10819f573d5b465a2dbfcf06a7899 Mon Sep 17 00:00:00 2001
From: Kye <kye@apacmediasolutions.com>
Date: Sun, 15 Oct 2023 01:53:05 -0400
Subject: [PATCH] exa error

---
 apps/paper_swarm.py                          | 66 ++++++++++++++++++++
 images/404.txt                               |  5 ++
 images/Swarms.md                             |  2 -
 pyproject.toml                               |  3 +-
 swarms/prompts/multi_modal_visual_prompts.py |  2 +-
 swarms/tools/autogpt.py                      |  1 -
 6 files changed, 73 insertions(+), 6 deletions(-)
 create mode 100644 apps/paper_swarm.py
 create mode 100644 images/404.txt
 delete mode 100644 images/Swarms.md

diff --git a/apps/paper_swarm.py b/apps/paper_swarm.py
new file mode 100644
index 00000000..203d79e4
--- /dev/null
+++ b/apps/paper_swarm.py
@@ -0,0 +1,66 @@
+"""
+Paper Swarm
+1. Scrape https://huggingface.co/papers for all papers, by search for all links on the paper with a /papers/, then clicks, gets the header, and then the abstract.
+and various links and then adds them to a txt file for each paper on https://huggingface.co/papers
+
+2. Feed prompts iteratively into Anthropic for summarizations + value score on impact, reliability, and novel, and other paper ranking mechanisms
+
+3. Store papers in a database with metadata. Agents can use retrieval
+
+4. Discord Bot // Twitter Bot
+"""
+
+
+import requests
+from bs4 import BeautifulSoup
+import os
+
+class Paper:
+    def __init__(self, title, date, authors, abstract):
+        self.title = title
+        self.date = date
+        self.authors = authors
+        self.abstract = abstract
+
+class Scraper:
+    def __init__(self, url):
+        self.url = url
+
+    def get_paper_links(self):
+        response = requests.get(self.url)
+        soup = BeautifulSoup(response.text, 'html.parser')
+        links = [a['href'] for a in soup.find_all('a', href=True) if '/papers/' in a['href']]
+        return links
+
+    def get_paper_details(self, link):
+        response = requests.get(self.url + link)
+        soup = BeautifulSoup(response.text, 'html.parser')
+        title = soup.find('h1').text
+        date_tag = soup.find('time')
+        date = date_tag.text if date_tag else 'Unknown'
+        authors = [author.text for author in soup.find_all('span', class_='author')]
+        abstract_tag = soup.find('div', class_='abstract')
+        abstract = abstract_tag.text if abstract_tag else 'Abstract not found'
+        return Paper(title, date, authors, abstract)
+
+class FileWriter:
+    def __init__(self, directory):
+        self.directory = directory
+
+    def write_paper(self, paper):
+        with open(os.path.join(self.directory, paper.title + '.txt'), 'w') as f:
+            f.write(f"h1: {paper.title}\n")
+            f.write(f"Published on {paper.date}\n")
+            f.write("Authors:\n")
+            for author in paper.authors:
+                f.write(f"{author}\n")
+            f.write("Abstract\n")
+            f.write(paper.abstract)
+
+scraper = Scraper('https://huggingface.co/papers')
+file_writer = FileWriter('images')
+
+links = scraper.get_paper_links()
+for link in links:
+    paper = scraper.get_paper_details(link)
+    file_writer.write_paper(paper)
\ No newline at end of file
diff --git a/images/404.txt b/images/404.txt
new file mode 100644
index 00000000..2fcf39ac
--- /dev/null
+++ b/images/404.txt
@@ -0,0 +1,5 @@
+h1: 404
+Published on Unknown
+Authors:
+Abstract
+Abstract not found
\ No newline at end of file
diff --git a/images/Swarms.md b/images/Swarms.md
deleted file mode 100644
index afd306fe..00000000
--- a/images/Swarms.md
+++ /dev/null
@@ -1,2 +0,0 @@
-# Agents
-* Agents are the fundamental building blocks of a swarm, they are indivi
\ No newline at end of file
diff --git a/pyproject.toml b/pyproject.toml
index 4ac04b5b..0480a343 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "swarms"
-version = "1.8.1"
+version = "1.8.2"
 description = "Swarms - Pytorch"
 license = "MIT"
 authors = ["Kye Gomez <kye@apac.ai>"]
@@ -46,7 +46,6 @@ redis = "*"
 Pillow = "*"
 chromadb = "*"
 agent-protocol = "*"
-exxa = "*"
 open-interpreter = "*"
 tabulate = "*"
 termcolor = "*"
diff --git a/swarms/prompts/multi_modal_visual_prompts.py b/swarms/prompts/multi_modal_visual_prompts.py
index 5bf2da2b..e6c70c1a 100644
--- a/swarms/prompts/multi_modal_visual_prompts.py
+++ b/swarms/prompts/multi_modal_visual_prompts.py
@@ -45,4 +45,4 @@ New input: {input}
 Since Worker Multi-Modal Agent is a text language model, Worker Multi-Modal Agent must use tools to observe images rather than imagination.
 The thoughts and observations are only visible for Worker Multi-Modal Agent, Worker Multi-Modal Agent should remember to repeat important information in the final response for Human.
 Thought: Do I need to use a tool? {agent_scratchpad} Let's think step by step.
-"""
\ No newline at end of file
+"""
diff --git a/swarms/tools/autogpt.py b/swarms/tools/autogpt.py
index d93211d1..80d280bc 100644
--- a/swarms/tools/autogpt.py
+++ b/swarms/tools/autogpt.py
@@ -146,7 +146,6 @@ class WebpageQATool(BaseTool):
         raise NotImplementedError
 
 
-
 @tool
 def VQAinference(self, inputs):
     """