Prismer.AI

Open Source OpenAI Prism Alternative

Learn · Paper Reading · Documentation · Roadmap · Demo Video

---

_{▶️ Watch Demo: Idea to Paper, End-to-End}

---

🚀 Live Products

Courses and learning resources for using Prismer effectively

AI-native PDF reader with citation graphs

---

What is Prismer.AI?

Prismer.AI is an open-source research platform that covers the entire academic workflow — from reading papers to publishing your own.

Current OSS Scope (This Repository)

This repository currently centers on:

• web/: Next.js research workspace (chat panel + multi-editor WindowViewer)
• docker/: OpenClaw-based container stack and gateway integration
• docs/: Architecture, runbook, contribution and roadmap documentation

Unlike tools that only handle writing (Overleaf) or note-taking (Notion), Prismer.AI integrates:

Feature	Description
📖 Paper Reading	AI-native PDF reader with citation graphs
💬 IM & Agent Protocol	Agent-to-agent messaging, groups, workspaces, real-time events
📄 Document Parsing	PDF/document parsing with markdown output
📊 Data Analysis	Jupyter notebooks with Python/R execution
✍️ Paper Writing	LaTeX editor with real-time preview
🔍 Citation Verification	Auto-checks references against academic databases
🤖 Multi-Agent System	Orchestrate specialized AI agents for research

---

Comparison

Feature	Prismer.AI	OpenAI Prism	Overleaf	Google Scholar
Paper Reading	✅	❌	❌	✅
Agent IM Protocol	✅	❌	❌	❌
Document Parsing	✅	❌	❌	❌
LaTeX Writing	✅	✅	✅	❌
Data Analysis	✅	❌	❌	❌
Code Execution	✅	❌	❌	❌
Citation Verification	✅	❌	❌	❌
Multi-Agent	✅	❌	❌	❌
Open Source	✅	❌	❌	❌
Self-Hosted	✅	❌	❌	❌

---

✨ Key Features

📖 Paper Reader

AI-native PDF reader for research papers with:

• Multi-document view with synchronized scrolling
• Bi-directional citation graph
• AI chat with paper context
• Figure/table extraction
• OCR data integration

✍️ LaTeX Editor

Modern LaTeX editor with:

• Real-time KaTeX preview
• Multi-file project support
• Template library (IEEE, ACM, Nature, arXiv)
• Smart error recovery with auto-fix

🔍 Citation Verification

LLMs fabricate citations. Prismer.AI solves this with a Reviewer Agent that validates every reference against academic databases (arXiv, Semantic Scholar, CrossRef) before it appears in your paper.

---

📦 Open Source Components

Component extraction is in progress. The @prismer/* entries below represent target packages being extracted from the monorepo:

Package	Version	Language	Description
@prismer/paper-reader		TypeScript	PDF reader with AI chat
@prismer/latex-editor		TypeScript	LaTeX editor with live preview
@prismer/academic-tools		TypeScript	arXiv, Semantic Scholar APIs
@prismer/jupyter-kernel		TypeScript	Browser-native notebooks
@prismer/code-sandbox		TypeScript	WebContainer code execution
@prismer/agent-protocol		TypeScript	Multi-agent orchestration

👉 See Component Documentation for usage examples.

---

🛠️ Self-Hosting

Quick Start

# 1. Clone & configure
git clone https://github.com/Prismer-AI/Prismer.git
cd Prismer
cp .env.example .env
# Edit .env — set OPENAI_API_KEY (any OpenAI-compatible API)

# 2. Start (Web + Agent containers)
cd docker
docker compose -f docker-compose.openclaw.yml up --build -d

Then open http://localhost:3000.

Service	URL	Description
Web UI	http://localhost:3000	Next.js frontend
Agent Gateway	http://localhost:16888	Container gateway (LaTeX, Jupyter, OpenClaw)
Health Check	http://localhost:16888/api/v1/health	Aggregated service health

See docker/README.md for detailed setup, configuration options, and API reference.

---

🗺️ Roadmap

Platform

Done	In Progress
✅ Paper Reader	🚧 Reviewer Agent
✅ IM API (agent messaging, groups, workspaces)	🚧 Documentation site
✅ LaTeX Editor	🚧 npm package extraction
✅ Jupyter Notebooks
✅ Multi-agent system
✅ Self-hosting (Docker)

Open Source Workspace

Done	In Progress	Planned
✅ Workspace code extraction & trim	🚧 Hardcoded URL cleanup	📋 Gateway compatibility layer
✅ Dead code cleanup (~9500 lines removed)	🚧 Lite Docker image (< 4GB)	📋 Local-mode SQLite persistence
✅ Static agent configuration	🚧 Quick Start guide	📋 Forkable CI pipeline
✅ LICENSE, CONTRIBUTING, SECURITY docs		📋 E2E tests for local mode
✅ K8s layer removal

See full roadmap and open-source optimization design for details.

---

🤝 Contributing

Contributions are welcome! Please read our Contributing Guide first.

---

⭐ Star Us

If you find Prismer.AI helpful, please consider giving us a star! It helps us grow and improve.

Star History

---

📄 License

• Components (@prismer/*): MIT License
• Workspace (web/): Apache-2.0 License
• Docker (docker/): Apache-2.0 License

---

_{Built for researchers, by researchers.}