EvoScientist
🔬 Harness Vibe Research with Self-evolving AI Scientists
EvoScientist is a self-evolving AI agent framework designed for end-to-end scientific research workflows. Built on LangChain and DeepAgents, it coordinates a team of six specialized sub-agents (planning, research, coding, debugging, analysis, writing) that autonomously explore topics, generate insights, and iteratively improve through self-distilled memory and auto-generated skills. The system operates in a human-on-the-loop paradigm, functioning as a research collaborator that learns scholarly taste over time while supporting multiple LLM providers (Anthropic, OpenAI, Google, MiniMax, NVIDIA) and interfaces (CLI, TUI, WebUI, messaging channels).
Key Features
Use Cases
- 01Automating literature reviews by scheduling daily arXiv paper summaries with the paper-navigator skill
- 02Running iterative machine learning experiments with autonomous code generation, debugging, and result analysis
- 03Conducting multi-day research projects where the agent builds domain knowledge through self-evolving memory
- 04Coordinating research tasks across messaging platforms while maintaining shared session state
- 05Generating and refining scientific papers with continuous verification and improvement cycles
- 06Building reusable research workflows through auto-generated skills from repeated patterns
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
EvoScientist — FAQ
What is EvoScientist?+
EvoScientist is a self-evolving AI agent system that autonomously conducts scientific research through a coordinated team of six specialized sub-agents. It builds persistent knowledge across sessions via self-distilled memory and can generate reusable skills from its own work patterns.
How do I install EvoScientist?+
Install via uv with 'uv tool install EvoScientist' or pip with 'pip install EvoScientist'. Requires Python 3.11+ (below 3.14). After installation, run 'EvoSci onboard' to configure API keys and preferences through the interactive wizard.
Which AI models and clients does it support?+
EvoScientist works with Anthropic Claude, OpenAI GPT, Google Gemini, MiniMax, and NVIDIA models. It provides its own interfaces (CLI, TUI, WebUI) and integrates with messaging platforms (Telegram, Slack, Feishu, WeChat, QQ) rather than plugging into external AI clients.
What API keys are required?+
You need at least one LLM provider key (ANTHROPIC_API_KEY, OPENAI_API_KEY, GOOGLE_API_KEY, MINIMAX_API_KEY, or NVIDIA_API_KEY). A TAVILY_API_KEY is optional for web search capabilities. OAuth sign-in is available for CLI coding agent subscribers.
Is EvoScientist free to use?+
The software is open-source under Apache 2.0 license and free to use. However, you'll incur costs from the LLM provider API calls based on your usage, especially with scheduled tasks.
How do scheduled tasks work?+
Use '/schedule add' with a cron expression to automate recurring research workflows. Tasks run unattended and can include optional acceptance checklists for quality validation. Manage schedules with '/schedule list', '/schedule pause', and '/schedule remove' commands.
How do I install EvoScientist?+
Open the source repository on GitHub and follow its README. EvoScientist is a agent — MCP Agents Market links you directly to the official repo.
Is EvoScientist free?+
EvoScientist is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.