This Obsidian plugin enhances knowledge management by integrating Large Language Models (LLMs) to automate the creation of structured, persistent knowledge bases. It targets Obsidian users seeking to transform raw notes into interconnected concepts, research summaries, and visual diagrams, thereby building a dynamic knowledge graph directly within their vault.

How It Works

Notemd acts as an Obsidian plugin, connecting to over 30 cloud and local LLM providers. It processes notes to automatically identify key concepts, generate [[wiki-links]], create new concept notes, summarize web research, translate content, and produce diagrams (Mermaid, Vega-Lite). Its core advantage lies in persisting AI-generated insights directly into the user's vault, unlike ephemeral chat-based tools, fostering an evolving, interconnected knowledge graph.

Quick Start & Requirements

Installation is recommended via the Obsidian Marketplace. Manual installation from GitHub releases is also supported. Configuration involves selecting an LLM provider in the plugin settings and providing necessary API keys or endpoint URLs. Users may need to run local LLM servers (e.g., Ollama, LMStudio) for local model integration. Official documentation and a FAQ are available.

Highlighted Details

• Extensive LLM Support: Integrates with numerous cloud (OpenAI, Anthropic, Google, Mistral, Groq) and local LLM providers, offering flexible model configuration per task.
• Knowledge Graph Automation: Features automatic wiki-linking, concept note generation with customizable paths, and pure concept extraction without modifying source files.
• Multi-modal Output: Generates Mermaid and Vega-Lite diagrams, supports translation between multiple languages, and can summarize web research fetched via Tavily or DuckDuckGo.
• Robustness & Stability: Includes retry logic for API calls, runtime transport fallbacks, and advanced error debugging modes for reliable processing, even with intermittent network issues or slow gateways.
• Developer Tools: Offers a developer mode with diagnostic panels for in-depth troubleshooting of LLM provider connections and API interactions.

Maintenance & Community

The project is maintained by Jacobinwwey. Contributions are welcomed via the GitHub repository. The latest commit date was June 6, 2026, indicating recent activity. Community links like Discord or Slack are not explicitly detailed in the README.

Licensing & Compatibility

The project is released under the MIT License, which permits broad use, including commercial applications and integration into closed-source projects.

Limitations & Caveats

Source-mutating operations like adding wiki-links are primarily limited to .md and .txt files; PDFs require pre-conversion. Web research with DuckDuckGo is experimental. Local LLM setups require running separate server applications and may encounter CORS issues. Diagram export for HTML artifacts is limited to iframe fallback previews.