LEARN: Lectures from Educators, Archive of Rounds & Notes

📚 CARDS: Crowdsourced Annotated Research for Debating Solutions

• 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

⚖️ FIAT: Forum for Issue Analysis on Topic

• 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

🎥 LEARN: Lectures from Educators, Archive of Rounds & Notes

• ~1,400 college NDT rounds: dating back to 1995, averaging 50+ new rounds per year, including recent TOC and NDT eliminations
• ~350 Public Forum rounds: (2015–present), ~125 Policy rounds: (2003–present), ~100 LD rounds (2019–present)
• ~1000 instructional videos: across 20 categories: topic lectures, camp coaching, kritik theory, counterplans, impact calc, novice intro, speaking & delivery, and more
• ~150 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

🔎 STREAM: Search with Top Result Extraction & Answer Model

• 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

📝 REASON: Research Editor for Annotated Summaries in Outline Notation

• Debate Editor ProseMirror engine, packing roughly 500 editing commands into ~30 thematic groups. Highlights (full 50+ feature list in the package README):
• Structured outline: pockets, hats, blocks, tags, cards, analytics, and undertags as first-class node types, with footnotes, tables, and live/transcluded zones
• Lossless .docx and .cmir round-trip: Verbatim Word interop plus a native gzip save format, including encrypted-.docx decryption and damaged-file salvage
• Bulk conversion & compression, automatic style cleanup on import, and a headless cardmirror-read CLI/MCP tool for AI-assistant access to files outside the app
• One-click structural styles: Pocket, Hat, Block, Tag, Analytic, Undertag, plus citation/underline/emphasis marks and acronym-aware variants
• Highlight, shading, and font-color pickers with standardization commands, a paintbrush mode, and highlight locking
• Card numbering, multiple condense modes, and a full editing-utilities set (shrink/regrow, short cites, live-zone refresh, heading move/copy/delete)
• Real-time collaboration (CRDT-backed via Loro) with share codes, invite links, and version recovery — plus an account-linked contacts list (requests, blocking, presence) to share a live card straight to a contact's account
• Speech-doc targeting: mark a doc as the live send target and send content at cursor or at end, with a persistent send history
• Flow integration: send cards or headings straight to a Flow column or cell, or pull content back
• Dropzone card exchange, Quick Cards, and a unified command-bar search across cards, commands, settings, and ~50 other site tools
• AI tools: ask-about-selection, AI-generated citations, translation, and AI-assisted text/formatting repair
• Flashcards with spaced-repetition review, a card cutter panel, voice dictation, and a reading-marker mode
• Runtime plugin registry, a 12-category menu bar, customizable keybindings, and per-user preferences synced to account settings

Features

Every tool below is live — the features page lists the same set, one entry per tool.

Research & Evidence

• Evidence Library — search shared, tagged evidence cards; Argument Library browses the same corpus by topic and case area.
• Team Research — topic coverage, prep room, review queue, task inbox, contributions feed and brainstorm board on one page.
• Research Workspace — the squad hub tying coverage, the library, quests, leaderboards and peer review together.
• Reason Editor and Debate Docs — outline-notation drafting with pockets, hats, blocks, tags and cards.
• File Market — sell research files and buy others'; the table of contents is free, the cards unlock on purchase.

Community & Progress

• Leaderboard, Contributor Awards, Daily Best Card and Progress — helpfulness score, tiers, badges and quest streaks.
• Daily Quests, Group Challenges and Debater Level — team goals and XP.
• LLM Card Scoring and Revision Incentives — score cards for quality, then earn reward for fixing them.
• Team Collaboration Mode — live prep notes grouped by topic, assignable to a teammate.
• Contacts and Notifications — share a live card with a teammate, and get pinged when a prep note is assigned to you.
• News & Forums — product updates, announcements and community threads.

Prep & Practice

• Prep Workspace — pre-round briefings, scout-to-strategy, opponent and judge profiles, prep notes, debater stats, practice partners and drills.
• Practice vs AI — debate an AI opponent in real turn order; the round simulator lives on the same page.
• AI Judge Decision — a decision grounded in the round's saved judge paradigm and flow summary.
• Speech Transcript Summaries, Argument Tree Outline and Flow Annotations — turn a round's flow into summaries, a filterable outline and timestamped notes.
• Word-Count Speeches — practise against a word budget instead of a clock.
• Prediction Markets — play-money bets on rounds, tournaments and team ratings.
• Speech Documents — every piece of evidence sent into your speech doc.

Coaching & Analytics

• Coach Workspace — AI Coach Mode, Coaching Programs, Team Calendar and Coach Materials as tabs.
• Team Rankings — TOC bid list alongside Elo ratings.
• Laptop-less Debating — prep, flow and speak from a phone.

Contributing

npx git0 debate/ai

• Start developing locally, develop features, open ideas in discussions, and submit a PR!
• API Reference — Interactive Scalar docs generated from debate-openapi.yml with debate-api-client (npm)
• Open questions and larger proposals belong in Discussions; bugs and scoped work belong in Issues.
• See LICENSE.md; third-party packages carry their own LICENSE and THIRD-PARTY-NOTICES.md.
• 🌟 Star this repo so it will grow and get new features!