Structure

Layers are inferred from where files sit in the tree, not from reading the code. A project that names its directories unconventionally will read oddly here — open the file browser to check anything the diagram implies.

Languages

Share of indexed source by file size. Binary and vendored files are excluded.

Feature verification

• Article summarization via LLMVerified

  Analyzes and provides a summary of the article the user entered

  Claimed on Devposthigh confidence

  • Backend/Backend.py:7 — analyze_website_LLM parses the article and sends it to together.Complete.create with a summarization prompt, returning the LLM output as 'main points'
  • Backend/index.py:48 — main3 calls Backend.analyze_website_LLM(str)['main points'] and wires it into the Taipy build_url task/output
• Hume.ai integration for text semantic analysisVerified

  We used Hume.ai to be able to analyze semantic in text

  Claimed on Devposthigh confidence

  • Backend/SentimentAnalysis.py:12 — imports HumeStreamClient and LanguageConfig from hume, connects with an API key, and sends article text via socket.send_text
• Python implementationVerified

  We coded the project in python

  Claimed on Devposthigh confidence

  • Backend/Main.py — all backend logic (Main.py, Backend.py, Parser.py, SentimentAnalysis.py, index.py) is written in Python
• Taipy backend/GUIVerified

  Taipy is used in the backend

  Claimed on Devposthigh confidence

  • Backend/index.py:1 — imports taipy, configures Core scenario/tasks/data nodes, and defines a Taipy Gui page with input/submit/text controls wired to the analysis functions
• Together.ai LLM integrationVerified

  We also used Together.ai for our LLM

  Claimed on Devposthigh confidence

  • Backend/Backend.py:2 — imports together, sets together.api_key, and calls together.Complete.create with model togethercomputer/llama-2-7b-chat
• User enters an article (URL) for analysisVerified

  Analyzes the article that the person entered

  Claimed on readmehigh confidence

  • Backend/index.py:101 — GUI exposes an <|{input_topic}|input|> field and submit button feeding into the scenario that runs sentiment and summary tasks on the given URL
  • Backend/Parser.py:12 — parse_website(website_url) fetches and extracts paragraph text from the entered URL via requests/BeautifulSoup
• Bias/misinterpretation detection via semantic sentiment analysisCode-supported

  The program is able to detect misinterpretation through semantic analysis / determines bias in language

  Claimed on Devpostmedium confidence

  • Backend/SentimentAnalysis.py:16 — get_sentiment_from_article uses Hume's LanguageConfig(sentiment={}) to score article text into 9 sentiment buckets
  • Backend/Main.py:9 — main() aggregates sentiment totals into a positive/negative/neutral label, but this measures sentiment/tone, not bias or 'misinterpretation' specifically; no distinct bias-detection logic exists
• Five key bullet points summaryCode-supported

  Summarizes the article by giving five key bullet points

  Claimed on Devpostlow confidence

  • Backend/Backend.py:14 — Prompt asks the LLM to 'summarize the following article in 5 sentence paragraph', not five bullet points; the output is a single text blob from the LLM, not a bulleted list, so the '5 bullet points' framing is only loosely implemented
• Type a concept and multiple related articles pop up with analysis (future work)Code-supported

  Next step is to be able to type in a concept and multiple articles popup with analysis and summaries

  Claimed on Devpostlow confidence

  • Backend/GoogleSearcher.py:9 — get_google_results(topic, amount) performs a Google search for a topic and returns multiple URLs, a plausible building block for this stated future feature, but it is not called from Main.py, Backend.py, or index.py, so it is not wired into the app

An AI agent derived these features from the project’s Devpost page and readme, then searched the code for each one. Verified features are backed by cited code; claimed-only features had no supporting code, which is not by itself proof a feature is missing.