Features

Everything a product mind needs.

Built for product managers, researchers, analysts, and anyone who uses AI to think and create.

🤖 AI Chat

Context-Aware AI Chat for User Stories & PRDs

Chat with Claude, GPT-4, Gemini, or local Ollama models. Every conversation is automatically saved to your project as a Markdown file — no manual copy-pasting needed, ever again.

  • ✓ Claude, OpenAI, Gemini, Ollama support
  • ✓ @file referencing — attach context from your project
  • ✓ Collapsible thinking process accordion
  • ✓ Auto-saved conversation history per project
ProductOS AI Chat interface showing multi-model selection and markdown-based conversation history
⚡ Workflows

Automate Competitive Research & Market Analysis

Build drag-and-drop workflows that run multi-step AI research automatically. Process competitor lists, analyze documents in parallel, and synthesize results — in minutes, not hours.

  • ✓ Drag-and-drop canvas workflow builder
  • ✓ Parallel processing — handle multiple items at once
  • ✓ Sequential & branching step types
  • ✓ 6-8 hours of work → 30 minutes
ProductOS Visual Workflow Canvas for automating multi-step AI research processes
🧠 Skills

Reusable AI agents trained for your tasks

Create specialized AI agent templates — Research Assistants, PRD Generators, Competitive Analysts — and reuse them across projects. Consistent quality, every time.

  • ✓ Custom agent roles & output formats
  • ✓ Reuse across unlimited projects
  • ✓ Share skills with your team
  • ✓ Skills marketplace & downloads
ProductOS Skills Library showing reusable AI agent templates for PM tasks
🗂️

Project Management

Organize all your research, chats, and notes by project. Full history always available, always searchable.

🔌

MCP Integrations

Connect GitHub, Jira, Notion, web search, databases — your AI gets real-time access to your tools.

🔐

AES-256 Encryption

API keys stored with military-grade encryption. No data sent to external servers. Your secrets stay yours.

📝

Markdown Native

All data stored as plain Markdown. Git-friendly, human-readable, portable. Zero vendor lock-in.

⚡

Native Performance

Built with Rust + Node.js. No browser lag, no bloat. Lightning-fast on macOS, Windows, and Linux.

🤝

Team Collaboration

Share projects, skills, and workflows with your team via Git. Works with any text editor or IDE.

Integrations

Bring your favorite AI. Connect any tool.

ProductOS works with every major AI provider and integrates with your existing stack via MCP.

AI Providers
MCP Tools
Technical Specs

Built for performance and privacy.

Specification Detail
Primary Stack Node.js Backend + React/TypeScript Frontend
Encryption AES-256-GCM (Military-grade for API keys)
Data Format Plain Markdown (.md) — Human readable and portable
Architecture Local-first, No cloud database required
AI Connectivity Native API (Claude, OpenAI, Gemini) + Local (Ollama) + MCP
Supported Platforms macOS, Windows, Linux
Common Questions

Everything you need to know.

Can't find the answer you're looking for? Check our documentation on GitHub.

Yes. ProductOS allows you to create reusable "Skills" (AI agents) specifically for PRD generation. By providing the AI with your project context, meeting notes, and research, it can generate a structured, high-quality PRD in seconds.

Absolutely. You can build workflows that take a feature description and automatically break it down into detailed user stories with acceptance criteria, formatted for Jira or Notion.

Yes. ProductOS is local-first. Your project files, API keys, and conversation history never leave your machine. It is designed specifically for PMs working with confidential strategy and R&D data.

All your data — including chats, research, and artifacts — is stored locally on your machine as human-readable Markdown files. We don't use external databases, so you have complete control and ownership of your research.

Yes! ProductOS integrates seamlessly with Ollama, allowing you to run powerful open-source models (like Llama 3 or Mistral) locally on your own hardware without sending data to the cloud.

Token Saver is a unique feature that uses smart history compaction. It reduces the size of your conversation context when sending it to AI models, potentially saving you up to 70% on API costs while maintaining accuracy.

Absolutely. ProductOS supports the Model Context Protocol (MCP), which lets you connect your AI to external tools like Jira, GitHub, Notion, and Slack for real-time data access and task automation.

Your API keys are stored locally in an encrypted vault using AES-256-GCM encryption. They are only decrypted when needed for requests and are protected by a master password you set during installation.

It's a comprehensive, chronological audit trail of every research step and AI interaction. It ensures your research is transparent, reproducible, and easy to trace back to the source.

Both. While it's a native desktop app, the Markdown-based storage makes it perfect for collaboration via Git. You can share projects, skills, and workflows with your team as easily as sharing code.

Yes! ProductOS is open source and built with transparency in mind. You can contribute, audit the code, or customize it to your needs on our GitHub repository.