Summary

InnoClaw is a self-hostable AI research workspace designed for grounded chat, paper study, and scientific research execution. It transforms server folders into AI-native environments, benefiting researchers, developers, and labs by providing cited answers over local files, reusable scientific skills, and a direct path from literature review to experimental execution.

How It Works

The system structures server folders as durable research workspaces, integrating file management, RAG-backed chat for grounded answers with citations, and a dedicated "Paper Workbench" for literature review. It supports reusable "Scientific Skills" and an "Execution" layer for orchestrating remote jobs and experiments. This multi-model friendly architecture allows integration with various AI providers like OpenAI, Anthropic, and Gemini, enabling a unified workflow from data indexing to research task completion.

Quick Start & Requirements

• Installation: Clone the repository, install Node.js 24+ (LTS recommended), run npm install, and start with npm run dev. Access via http://localhost:3000.
• Prerequisites: Requires at least one configured AI provider (e.g., OpenAI, Anthropic, Gemini).
• Documentation: Detailed setup and usage guides are available at https://SpectrAI-Initiative.github.io/InnoClaw/ and in docs/getting-started/installation.md.

Highlighted Details

• Recent updates (March 2026) include dynamic model discovery, per-model base URL routing, runtime tool-calling toggles, Node.js 24+ support, and multimodal LLM integration for paper analysis.
• Features encompass GitHub Skills Import Preview, Obsidian note export, remote HPC/SLURM execution, Kubernetes cluster configuration, and a deep research module with an execution pipeline.
• Supports importing and utilizing a claimed 206 SCP scientific skills across domains like drug discovery and genomics.
• Multimodal vision capabilities allow analysis of PDF images (figures, tables) within paper study and discussion workflows.

Maintenance & Community

Active development is evident with frequent updates in March 2026. Community support and discussion are available via a Feishu group (link in docs/README_CN.md). Bug reports and feature requests can be submitted via GitHub Issues.

Licensing & Compatibility

The project is licensed under the Apache-2.0 license, which is permissive and generally suitable for commercial use and integration into closed-source projects.

Limitations & Caveats

As a self-hostable application, InnoClaw requires user-managed infrastructure and configuration of AI providers. The Node.js 24+ runtime is a specific dependency. While extensive, the "206 SCP scientific skills" may require specific setup or be a subset of available functionalities.