Plan
Ask clarifying questions, gather context, break tasks into structured plans. No execution until you're ready.
OpenMLRML Research Agent
Plans tasks, researches papers, writes drafts, and executes code — end to end, in one conversation.
Skip to content
Quick Start
bash
git clone https://github.com/xprilion/OpenMLR.git
cd OpenMLR
cp .env.example .env
make upOpen http://localhost:3000. Create an account. Configure API keys in Settings > Providers.
No API keys needed to start — the app guides you through setup after login.
One-Click Deploy
See Quick Start for all deployment options or Setup & Installation for detailed instructions.
How It Works
OpenMLR uses two modes to keep the agent focused:
- Plan mode (P) — The agent asks questions, gathers context, and creates structured plans. No code execution, no file writes. Messages have an amber border.
- Execute mode (E) — The agent does the work: researches papers, writes drafts, runs experiments. All tools available. Messages have a blue border.
Switch modes with the P/E button in the input area or press Cmd+M (Mac) / Ctrl+M (Windows/Linux) to toggle.
Key Features
- Paper research — OpenAlex, Semantic Scholar, arXiv, CrossRef, Papers With Code. Full paper reading, citation graphs.
- Paper writing — Section-by-section drafting with auto-save. Preview + export (Markdown/LaTeX) in the Paper tab.
- MCP servers — Connect remote HTTP/HTTPS MCP servers with custom auth. Per-server mode config, live connection status, @ mentions.
- @ mentions — Type
@in the chat to reference MCP servers or workspace files. The agent uses its tools to interact with the referenced resources. - Sub-agent streaming — Research tool spawns independent agents with nested tool call visibility.
- Background jobs — Celery + Redis. Close the browser, come back later.
- Per-conversation parallelism — Multiple conversations process simultaneously.
- Onboarding flow — Guided setup when no LLM provider is configured.