Ar9av/obsidian-wiki

Framework for AI agents to build and maintain a digital brain through Obsidian wiki | Memory System for Agents

What it solves

It prevents the loss of knowledge gained during AI agent interactions. Instead of letting valuable insights disappear into fragmented chat logs or repeatedly running expensive RAG processes, this tool allows an AI agent to compile and maintain a persistent, interconnected "digital brain" of markdown files that the user owns and can read independently.

How it works

The system turns a folder of markdown files (an Obsidian vault) into a long-term memory for AI agents. It uses a set of "skills" (slash commands) that agents can execute to ingest data from various sources—such as PDFs, URLs, and chat exports—and distill that information into a structured wiki.

Key mechanisms include:

  • Knowledge Compilation: New information is merged into existing pages rather than just accumulated, with contradictions flagged and duplicates removed.
  • Session Lifecycle: It automatically injects relevant context (user profiles, open threads) at the start of a session and captures reusable knowledge before a session ends.
  • Local-First Storage: Everything is stored as plain markdown on disk, requiring no API keys, embedding models, or vector databases for the core vault functionality.
  • Graph-Aware Queries: It provides tools for the agent to query the structural relationships between notes, making complex architectural questions faster and more accurate than standard grep-based searches.

Who it’s for

Developers and power users who use AI agents (like Claude Code, Cursor, or Windsurf) and want a persistent, portable, and human-readable knowledge base that evolves alongside their projects.

Highlights

  • Agent-Agnostic: Compatible with a wide range of agents including Claude Code, Cursor, Codex, and Gemini CLI.
  • Local Ownership: Data is stored in plain markdown, making it greppable, diffable, and independent of any vendor lock-in.
  • Comprehensive Toolset: Includes 39 skills for ingesting, querying, linting, and deduplicating knowledge.
  • Session Persistence: Maintains continuity across different AI sessions by tracking who the user is and what is currently active.
  • Python API: Allows developers to integrate vault memory directly into Python applications.

Related

  • Project
    huytieu/COG-second-brain

    An agentic second brain that uses AI agents and local Markdown files to automate knowledge synthesis, professional workflows, and personal organization.

  • Project
    undefined-ui/second-brain-os

    An open‑source guide and starter kit that turns an Obsidian vault into a personal “second brain” automatically curated by a Claude Code LLM agent. It provides a vault template, 18 LLM‑driven skills, 72 slash commands, scheduled sub‑agents, and detailed docs covering setup, ingestion, graph handling, and maintenance.

  • Project
    mindmuxai/brain.md

    A persistent, file-based memory layer for coding agents that stores project decisions and constraints as plain Markdown in the repository.

  • Project
    itechmeat/open-second-brain

    A memory layer for AI agents that stores knowledge as Markdown files in an Obsidian vault, enabling persistent learning and human-editable memory.

  • Project
    akitaonrails/ai-memory

    ai‑memory is a Rust‑based, self‑contained service that gives AI coding assistants a shared, long‑term, markdown‑backed memory. It automatically captures prompts and tool calls via hooks, stores them as git‑versioned markdown pages, and lets any supported agent (Claude Code, Codex, Cursor, Gemini CLI, etc.) retrieve a concise hand‑off on a different machine or after a teammate’s session. By default it works offline with zero LLM calls; optional LLM‑driven “dream” consolidation, vector search, and aging policies are configurable. The project provides native packages, Docker images, and a macOS menu‑bar app, and includes built‑in multi‑user auth, audit logs, and per‑project sharing.