[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:arkon":3},"\u003Ch1>Arkon - The Open-Source Enterprise AI Knowledge Hub & MCP Server\u003C\u002Fh1>\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fnduckmink\u002Farkon\u002Fblob\u002FHEAD\u002FLICENSE\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-PolyForm_Internal_Use_1.0.0-blue.svg\" alt=\"License\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fnduckmink\u002Farkon\u002Fstargazers\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnduckmink\u002Farkon.svg\" alt=\"GitHub Stars\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fwww.docker.com\u002F\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDocker-Ready-blue.svg\" alt=\"Docker\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fnduckmink\u002Farkon\u002FHEAD\u002Fdocs\u002Fassets\u002Farkon.png\" alt=\"Arkon\" width=\"720\" \u002F>\n\u003C\u002Fp>\u003Cp>\u003Cstrong>Arkon\u003C\u002Fstrong> is a self-hosted, enterprise-grade knowledge management layer that bridges organizational data and AI clients. It runs as a centralized \u003Cstrong>MCP Server\u003C\u002Fstrong> (Model Context Protocol), compiling your SOPs, policies, and internal docs into a structured, traceable knowledge wiki - then serving that wiki to Claude and other LLMs through a single permission-scoped endpoint.\u003C\u002Fp>\n\u003Ch2>🚀 Why Arkon?\u003C\u002Fh2>\n\u003Cp>In most organizations, AI adoption is fragmented. Employees copy-paste documents into chatbots, producing inconsistent context, security risks, and duplicated work.\u003C\u002Fp>\n\u003Cp>\u003Cstrong>Arkon treats AI as a managed organizational resource.\u003C\u002Fstrong> Every employee gets the right context, automatically and securely - filtered by their department and role.\u003C\u002Fp>\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fnduckmink\u002Farkon\u002FHEAD\u002Fdocs\u002Fassets\u002Fhow-arkon-works.png\" alt=\"How Arkon Works\" width=\"720\" \u002F>\n\u003C\u002Fp>\u003Chr \u002F>\n\u003Ch2>✨ Key Features\u003C\u002Fh2>\n\u003Ch3>🧠 Intelligent Knowledge Wiki - the MRP Pipeline\u003C\u002Fh3>\n\u003Cp>Unlike a vector database that just chunks and indexes, Arkon's \u003Cstrong>MRP pipeline\u003C\u002Fstrong> (\u003Cstrong>M\u003C\u002Fstrong>ap → \u003Cstrong>R\u003C\u002Fstrong>educe → \u003Cstrong>P\u003C\u002Fstrong>lan-review → \u003Cstrong>R\u003C\u002Fstrong>efine → \u003Cstrong>V\u003C\u002Fstrong>erify → Commit) actually compiles documents into a coherent wiki of interlinked pages.\u003C\u002Fp>\n\u003Cul>\n\u003Cli>\u003Cstrong>Plan review before write:\u003C\u002Fstrong> every ingestion produces a human-reviewable plan listing which wiki pages will be created or updated. Editors can regenerate the plan with feedback before any page is written.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Page merge instead of overwrite:\u003C\u002Fstrong> when a new source touches an existing wiki page, content is LLM-merged so prior knowledge is never lost.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Traceable claims:\u003C\u002Fstrong> every page records the source documents it was compiled from.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Image-aware:\u003C\u002Fstrong> vision captions are baked into source text before compilation, so wiki pages reference the right images in the right places.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Resumable:\u003C\u002Fstrong> drafts persist mid-pipeline; a crashed run resumes without re-doing the expensive LLM work.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch3>📚 Wiki Browser & Knowledge Graph\u003C\u002Fh3>\n\u003Cul>\n\u003Cli>Three-panel layout: page tree, content, backlinks & outlinks.\u003C\u002Fli>\n\u003Cli>Full-text + semantic (pgvector) search.\u003C\u002Fli>\n\u003Cli>Interactive knowledge graph visualization (per-scope or global).\u003C\u002Fli>\n\u003Cli>Wikilink-style cross-references between pages.\u003C\u002Fli>\n\u003Cli>Version history and rollback on every page.\u003C\u002Fli>\n\u003Cli>Draft proposal → editor review → approval workflow.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fnduckmink\u002Farkon\u002FHEAD\u002Fdocs\u002Fassets\u002Farkon-wiki-ui.png\" alt=\"Arkon Wiki Browser UI — three-panel knowledge base with page tree, content viewer, and knowledge graph\" width=\"720\" \u002F>\n\u003C\u002Fp>\u003Ch3>🏢 Department & Global Scopes\u003C\u002Fh3>\n\u003Cp>Precise context boundaries with their own scoped wiki, document set, and department rosters.\u003C\u002Fp>\n\u003Cul>\n\u003Cli>\u003Cstrong>Department-level isolation\u003C\u002Fstrong> for HR, Legal, Engineering, etc.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Global scope\u003C\u002Fstrong> for company-wide SOPs, policies, and general knowledge.\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Hard scope enforcement:\u003C\u002Fstrong> members only see knowledge from their assigned departments or the global realm - at the API, MCP, and search layers.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch3>🛂 Fine-Grained RBAC\u003C\u002Fh3>\n\u003Cp>Role-based access control at the department and global levels.\u003C\u002Fp>\n\u003Cul>\n\u003Cli>Built-in roles: \u003Cstrong>Viewer · Contributor · Editor · Admin\u003C\u002Fstrong> to control draft, edit, and setting access.\u003C\u002Fli>\n\u003Cli>Granular permissions (\u003Ccode>doc:read:own_dept\u003C\u002Fcode>, \u003Ccode>wiki:edit:all\u003C\u002Fcode>, \u003Ccode>org:settings:manage\u003C\u002Fcode>, ...).\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Audit log\u003C\u002Fstrong> for every privileged action - settings changes, plan approvals, and role updates.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch3>🔌 MCP Server for Claude & Other AI Clients\u003C\u002Fh3>\n\u003Cp>Employees connect Claude Desktop or Claude.ai to Arkon via \u003Cstrong>OAuth 2.1 + PKCE\u003C\u002Fstrong> — just add the server URL and sign in through the browser. No manual token copying required. The MCP server exposes:\u003C\u002Fp>\n\u003Cul>\n\u003Cli>**Wiki too\u003C\u002Fli>\n\u003C\u002Ful>\n",1786232074676]