Logo
Deploy Now

SaaS Alternative

Notion Confluence

Categories

Self-Hosted AI LLM Knowledge Base Documentation Wiki MCP Note-Taking

Stars

590

Forks

97

Watchers

4

Developer links

Last Updated: 08-24-2026 05:56 AM (UTC)

Sage Wiki

Sage Wiki turns a pile of unstructured documents into a fully interlinked, searchable wiki by running them through a five-pass LLM compiler pipeline. Inspired by Andrej Karpathy's vision of LLM-compiled knowledge bases, the pipeline processes source files through diff detection, summarization, concept extraction, image captioning, and cross-reference discovery, with parallel LLM calls and checkpoint/resume for vaults scaling to 100,000+ documents. The typed ontology graph stores entities and relations with BFS traversal, configurable relation types, multilingual synonyms, and a promotion/demotion lifecycle backed by grounding verification and consensus scoring. Multi-format ingestion handles Markdown, PDF, Word, Excel, PowerPoint, EPUB, email, CSV, images, and code files without manual tagging. LLM provider support spans Anthropic, OpenAI, Gemini, Ollama, and any OpenAI-compatible API, with per-pass model routing enabling cost optimization by assigning cheaper models to simpler tasks. The built-in MCP server exposes 17 tools over SSE transport for integration with Claude, Cursor, and any MCP-compatible agent, while native Obsidian vault overlay ensures existing note workflows remain undisrupted. Team deployment supports Git-synced shared wikis, centralized server access, and hub federation across multiple projects. Ships as a single Go binary with Docker Compose multi-arch images serving the web UI on port 3333. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Sage Wiki
Sage Wiki
Sage Wiki
Sage Wiki
Sage Wiki

Benefits

  • Automatic Knowledge Compilation
  • Five-pass LLM pipeline automatically extracts concepts, discovers cross-references, generates summaries, and builds an interlinked ontology graph from unstructured documents without manual tagging.
  • Universal Document Ingestion
  • Processes Markdown, PDF, Word, Excel, PowerPoint, EPUB, email, CSV, images, and code files through a single pipeline scaling to 100,000+ documents with checkpoint/resume.
  • Multi-Provider LLM Flexibility
  • Supports Anthropic, OpenAI, Gemini, Ollama, and any OpenAI-compatible API with per-pass model routing for cost optimization and local model fallback without internet dependency.
  • MCP Agent Memory Layer
  • Built-in MCP server with 17 tools over SSE transport enables Claude, Cursor, and compatible agents to read, query, and contribute to the wiki as persistent institutional memory.

Features

  • Ontology Graph Engine
  • Typed entity-relation graph with BFS traversal, configurable relations, multilingual synonyms, and grounding verification with consensus-based promotion lifecycle.
  • Semantic Search
  • Chunk indexing with query expansion, re-ranking, and graph expansion delivers contextual search results across the full knowledge base without external search services.
  • Obsidian Integration
  • Native vault overlay mode opens compiled wikis directly in Obsidian with bidirectional linking, preserving existing note workflows and enabling offline editing.
  • Team Federation
  • Three deployment patterns support Git-synced shared wikis, centralized server with web UI access, and hub federation for multi-project organizations up to 50 users.
  • Docker Multi-Arch
  • Published to GHCR and Docker Hub with amd64 and arm64 images, serving the web UI on port 3333 with Syncthing-based cross-device synchronization.