Obsidian plugins with #knowledge-graph#

The Karpathy LLM Wiki plugin turns notes into an interlinked wiki and lets you query that graph from a side panel inside the editor. It extracts entities and concepts into separate pages, adds aliases, builds bidirectional wiki links and maintains an index, while keeping source notes unchanged. Retrieval uses a five stage cascade with lexical matching, LLM generated keywords, local substring scanning, fallback reseeding and Personalized PageRank over wiki links instead of embeddings or a vector database. The plugin also includes linting, duplicate and contradiction checks, smart repair flows, batch ingestion, PDF ingestion with provider specific and local OCR paths and support for many cloud and local model providers. Query answers link back into the wiki and PDF markdown can be cached under .obsidian/plugins/karpathywiki/pdf-cache/.

The InfraNodus AI Graph View plugin transforms Obsidian vaults into dynamic knowledge graphs, enabling users to visualize and analyze connections between ideas, concepts, and topics. It leverages advanced clustering and network science to highlight key topics, identify gaps, and generate research questions or ideas using built-in AI. The plugin can analyze single or multiple files, include backlinks and unlinked mentions, and provide rich graph analytics such as centrality, modularity, and influence metrics.

The Related Notes plugin enhances knowledge discovery in your Obsidian vault by automatically surfacing contextually similar notes based on their content. Using advanced similarity algorithms, it identifies relevant connections between notes and presents them in a dedicated pane, allowing users to explore and link related concepts with a single click. It operates entirely offline, ensuring privacy while handling both small and large vaults efficiently through adaptive indexing and hybrid search methods. The plugin intelligently prioritizes frequently accessed and newly created notes for indexing and supports real-time similarity detection for others. Visual indicators and structured link insertion streamline the linking process without disrupting your note organization.

The Neural Composer plugin adds graph based AI chat to your vault by running a LightRAG server, indexing notes into a knowledge graph and combining graph traversal with vector search for more connected answers. It can start and stop the backend automatically, watch a folder for reindexing, show processing status in the file explorer and open a 2D or 3D graph view for exploration. Answers include citations linked to the source notes and text chunks, and the plugin can expose graph tools through MCP clients. It also supports remote LightRAG servers, multiple model providers and a fully local setup with Ollama, while keeping API keys in the local vault data and avoiding runtime telemetry.

The Auto Keyword Linker plugin automatically converts predefined keywords into wiki-style links throughout your vault, building a richly interconnected knowledge graph without manual bracket typing. Define keywords once with their variations like "ML", "machine learning", "neural networks" and the plugin intelligently finds and links them while respecting existing links, code blocks, and markdown formatting. AI-powered suggestions analyze your vault to discover frequently used terms you might have missed. Advanced features include folder specific scoping for context aware linking, block references that jump to specific sections, relative path support and automatic tag management. Preview changes before applying them, enable auto linking on save for seamless workflow or process your entire vault in bulk. Import and export keyword configurations for backup or team sharing.

The Semantic Canvas plugin for Obsidian enhances the use of canvases by allowing users to visually edit file properties. It enables bulk property updates across multiple markdown files linked within a canvas, as well as the creation of semantic links between notes. The plugin provides intuitive tools for managing properties based on group membership and link relationships. Users can create knowledge graphs, perform mass edits, and even represent diagrams like Venn diagrams or Kanban boards. This plugin facilitates a more interactive and visual approach to working with notes and their properties, making it ideal for organizing complex information and improving note relationships.

The Connections plugin lets you create and visualize named relationships between notes in a dedicated sidebar pane. Define custom connection types like "is married to" or "is the father of" and link notes together with meaningful context. Connections can be added manually through a command or automatically using mapped connection types based on frontmatter properties. The plugin supports flexible grammar with configurable subjects, so you can control whether the source or target note appears first in the relationship description. Perfect for building knowledge graphs with explicit semantic relationships. Works alongside the Folder Notes plugin.

The Discourse Graphs plugin turns research notes into structured discourse objects so questions, claims, evidence and sources can be created, linked and revisited as part of a larger argument. It lets you define custom node types, relation types and saved relations, then use those structures inside notes or on a dedicated canvas view for mapping connections. Text can be converted into nodes from a shortcut, context menu or command palette, while relations can be added from a sidebar panel or drawn directly between cards on a canvas. This makes it suited to academic workflows where people need clearer reasoning trails, reusable evidence chains and a more deliberate way to build knowledge over time.