Being is Becoming —
Dream Emerges Destiny.
Whatever Research Can Be —
That is What It Must Become.
If AI is Humanity's Last Invention —
Collective Consciousness is the Final Frontier.
- 🖍️ Auto-Highlight Agents: agents highlight and underline as many words as needed on a slider, multiple options
- 🔎 Full-text search: across thousands of tagged, annotated evidence cards spanning policy, LD, PF, and college formats
- 🤖 AI-powered annotation: one-click summaries, warrant extensions, and logic-flaw detection per card
- 📖 Three reading modes: plain text, highlighted tags, and underline-only for fast cutting
- 📎 Citation auto-formatter: and one-click flow integration — paste directly into your speech doc
- 📱 Mobile-responsive: with full-screen card overlays for reading on the go
- 📝 Auto-Research Outlines: agents outline the topic to keyphrases and monitor for new quotes
- 🧠 Recommendation Agents: AI assists with research, summarization, flaw detection, and comparative quote analysis
- 🧑⚖️ Judge Decision: agents prompts recommend multiple judge decision options, speech to flow, quote to response options
- 📊 Multi-column flow spreadsheet: format-specific speech columns for PF, LD, Policy, and NDT with inline editing
- 🔗 Shareable round URLs: every round gets a permanent link; share with judges or teammates instantly
- 🏆 Round management: tournament setup, team pairing, judge assignments, and round notes in one place
- 📄 Speech docs: full markdown editor per speech with email sharing to judges and coaches
- ⏱️ Smart timers: format-aware prep and speech timers with audio/visual alerts and auto-advance
- 👥 Collaboration: invite judges and spectators by email; view-only and edit roles supported
- 🗄️ Archive system: save, browse, and restore any past round with full flow history
- 📲 Mobile-optimized: responsive design with swipe navigation between speech columns
- 🎓 ~1,400 college NDT rounds: dating back to 1995, averaging 50+ new rounds per year, including every recent TOC and NDT elimination round
- 🗣️ ~350 Public Forum rounds: (2015–present), ~125 Policy rounds: (2003–present), ~90 LD rounds (2019–present)
- 📺 ~900 instructional videos: across 20 categories: topic lectures, camp coaching, kritik theory, counterplans, impact calc, novice intro, speaking & delivery, and more
- ⭐ ~100 hand-curated top picks: the highest-value rounds and lectures selected for study
- 🔍 Searchable grid with filter by title, channel, year, or view count; inline YouTube playback with thumbnails
- 📘 200-term Debate Dictionary: with plain-English definitions for theory, kritik, and procedural jargon
- 🏅 26 years of national champion records: (2000–2025) across NDT, Policy, LD, and PF
- 📈 Team Rankings: TOC bid list + DebateDrills Elo dual ranking system
- 🔍 Web Search: 70+ popular sites search across 10 categories: Web Search, Academic, Videos, Images, Files, News, etc
- 📝 Article Preview: Extract, format with APA cite, and summarize articles, PDFs, Youtube, and URLs before reading them
- 🤖 User Choice of LLM: OpenAI, Claude, Gemini, Groq, Ollama, Anthropic, etc
- 📄 File Upload Support: Ask questions about PDFs, URLs, DOCX, Google Docs, and Youtube
- 📚 Search History: All searches saved with memories, except for privacy mode
- ❓ Follow-up Questions: Generate follow-up questions to ask language models
- 📝 Complex Rich Text Editor: full featured alternative to Google Docs based on Meta's Lexical with core features and fast ease of use
- 📂 Nested Document Tree: organize research notes with a nested document organizer with drag-and-drop, tabs, and custom storage sources
- 🖱️ Context Menu: right-click to access quick actions for seamless document management
- 🔍 Full-Text Search: instantly find documents by title or content with full-text search
- ✨ AI Rewriting: leverage AI to rewrite and improve your text directly within the editor
- 👥 Team Management: collaborate with team members and manage access rights
- 🔄 View Modes: switch between Formatted, parsing HTML, and Markdown views for versatile editing
- 🛠️ Find & Replace: powerful search and replace functionality with match highlighting
- 📥 Google Docs Integration: seamless export, import, and sharing capabilities
- 💾 Persistent Storage: reliable SQLite storage ensures your data is safe and accessible
- ⌨️ Keyboard Navigation: efficient keyboard shortcuts for power users
- 💬 Research Quotes: capture and organize key quotes and insights from your research
- The Debate Singularity is Happening. AI systems can already assist with research, summarization, flaw detection, and comparative argument analysis, and prior debate-focused systems have shown that computational tools can support live argumentative tasks (Lippi & Torroni, n.d.; Association for Computational Linguistics, 2023). In this paper, the “singularity” does not mean full automation of judgment; it means the rapid convergence of debate practice, knowledge organization, and machine-assisted reasoning into one workflow.
- Collective Thought Engine. Crowdsourced research can be represented as a weighted argument space where claims gain influence through reuse, support, contestation, and contextual relevance (Bikakis et al., 2023). This turns debate from a sequence of isolated rounds into a shared reasoning substrate in which public arguments can be compared at scale.
- Transparent Reasoning. Argumentative AI is most useful when users can inspect the exact sentences, citations, and inferential steps behind an output, a principle aligned with work on contestable human-AI decision-making and explainability (Xiao & Greer, 2023). A debate-native system should therefore expose not just conclusions, but the evidence path that produced them.
- Outcome Simulation Trees. Users should be able to model likely responses, counterarguments, and downstream consequences across multiple branches of a controversy (Homer-Dixon & Karapin, 1989). This extends debate preparation into a formal simulation environment for testing which lines of reasoning remain persuasive across audiences and contexts.
- Outlines of Current News Issues. The most practical use case is live news outlines. Each article, quote, or claim becomes part of a topic tree, allowing users to track how a story evolves, where the main disagreements are, and which warrants support each side.
- Solving Post-Self Alignment. Modern discourse is fragmented by platform incentives, ideological sorting, and partial information environments. A shared debate outline can function as a corrective by placing opposed claims into one visible structure, making disagreement legible without reducing it to caricature.
- Topic Research Unified Tree Hierarchy (TRUTH). We call the resulting structure the Topic Research Unified Tree Hierarchy, or TRUTH: a hierarchical representation of issues, claims, evidence, and value conflicts. TRUTH is designed to help models identify overstatement, missing warrants, and unsupported leaps while grounding outputs in a common research map.
Start developing locally, develop features, open ideas in discussions, and submit PR!
npx git0 debate/ai