LibreAI / Productivité / obsidian-second-brain
// productivity · eugeniughelbur/obsidian-second-brain

obsidian-second-brain

obsidian-second-brain est un projet de productivité open source, créé par eugeniughelbur (eugeniughelbur/obsidian-second-brain sur GitHub). Il compte 4 697 étoiles sur GitHub et 580 forks, il est publié sous la licence MIT et a été mis à jour pour la dernière fois sur GitHub le 7 octobre 2026.

Persistent memory for Claude Code and other CLI agents, stored as plain Markdown in your Obsidian vault

★ 4,7 k étoiles⑂ 580 forksMITmis à jour oct. 2026
▦ visuel — le README de ce projet ne contient aucune capture intégrable,
voici donc un emplacement réservé. Le guide ci-dessous est complet malgré tout.
// détails du projet

Détails du projet

Étoiles GitHub
4 697
Forks
580
Licence
MIT
Dernière mise à jour sur GitHub
7 octobre 2026
Catégorie
Productivité
Type de fiche
Guide d'installation complet
Site officiel
Official site
// fonctionnement

Comment ça marche

  1. Prepare your vault

    Use an existing Obsidian vault or create a fresh one to serve as the memory store.

  2. Clone and install the skill

    Clone the repo and follow its install instructions for your agent (Claude Code first-class; other CLIs supported).

  3. Initialize

    Run the init command to set up the vault structure and pick a role preset (developer, founder/operator, writer, or researcher).

  4. Save and recall

    Capture a conversation with the save command and recall it in a later session with the find command.

// fonctionnalités

Pourquoi on l'adore

45+ slash commands

Commands for saving, ingesting, finding, synthesizing, reconciling, daily notes, tasks, people, calendar, and more.

Self-rewriting notes

New sources update existing pages instead of appending duplicates; contradictions are reconciled automatically.

Ingest anything

URLs, PDFs, audio (transcribed locally with Whisper), and screenshots are distributed across entity pages, task boards, and daily notes.

Scheduled agents

Four scheduled agents — morning brief, nightly consolidation, weekly review, and vault-health check — maintain the vault unattended.

Key-less research toolkit

Commands pull from X, the web, and YouTube with cited dossiers; vault-first deep research fills only the gaps in what you already know.

Plain-Markdown, portable memory

Memory lives as Markdown in your Obsidian vault and can be exported as a snapshot any AI tool can read.

Advertisement
// lancez-le

Lancez-le vous-même

  1. Clone the repository

    Clone the project and enter its directory.

    git clone https://github.com/eugeniughelbur/obsidian-second-brain.git
    cd obsidian-second-brain
  2. Install for your agent

    Follow the README's install section for Claude Code or your CLI agent, so the /obsidian-* commands become available.

    # see README install section for the exact command for your agent
  3. Initialize and save

    Inside your agent, initialize the vault, then capture and recall memory.

    /obsidian-init
    /obsidian-save
    /obsidian-find
// cas d'usage

Ce que vous pouvez créer

  • Stop re-briefing Claude Code on your project at the start of every session
  • Turn meeting recordings and articles into cross-linked notes automatically
  • Have a nightly agent reconcile contradictions and synthesize patterns in your notes
  • Research a topic with citations and file the dossier straight into your vault
Advertisement
// faq

Questions fréquentes

Who maintains obsidian-second-brain?

Eugeniu Ghelbur, an AI automation engineer; the project builds on Andrej Karpathy's LLM Wiki pattern.

Is it free?

Yes, MIT-licensed. You use it with your existing agent subscriptions; the README emphasizes key-less web research, and transcription uses a local Whisper install.

Does it need a paid API key?

No separate key is required by the skill itself — it runs inside agents you already have (such as Claude Code), and several research commands are described as key-less by default.

How does it compare to Obsidian's own AI plugins?

Those typically add chat or search inside Obsidian. This skill makes the vault the agent's long-term memory across CLI tools, with self-rewriting pages and scheduled maintenance agents.

How do I get started?

Clone the repo, install it for your agent per the README, run `/obsidian-init`, then try `/obsidian-save` and `/obsidian-find`.

// continuez l'exploration

Outils similaires

Productivité

obsidian-mind

A self-organizing Obsidian vault that gives AI coding agents persistent memory across sessions.

★ 5,0 k étoilesMIT
Voir le guide →
Productivité

jarvis-dashboard

An Obsidian dashboard for your Claude Code sessions.

★ 102 étoilesMIT
Voir le guide →
Productivité

OB1

Open Brain (OB1): one database, one AI gateway, one capture channel — shared memory for every AI you use

★ 4,7 k étoilesNOASSERTION
Voir le guide →
Productivité

claude-for-legal

Anthropic's plugin suite for legal workflows — practice-area plugins, named agents and MCP connectors.

★ 9,6 k étoilesApache-2.0
Voir le guide →
Productivité

claude-content-writer

A Claude Code skill that writes content in your voice.

★ 52 étoilesCustom
Voir le guide →
Productivité

pm-ba-claude-skills

5 no-cost Claude AI skills for Product Managers & Business Analysts — PRDs.

★ 51 étoilesMIT
Voir le guide →
// alternatives

Alternatives en Productivité

Les autres projets de la catégorie Productivité avec le plus d'étoiles du catalogue LibreAI — la sélection réaliste d'alternatives si obsidian-second-brain ne vous convient pas.

  • worldmonitor★ 88,1 k

    Real-time global intelligence dashboard: AI news aggregation, geopolitics and infrastructure tracking

  • Memos★ 63,9 k

    Memos — a personal timeline for quick notes: capture short memos in seconds, find them later by search, tag or date; open source and self-hosted

  • Joplin★ 56,8 k

    Joplin — the privacy-focused note-taking app: Markdown notes and to-dos on every platform, offline-first, with end-to-end encrypted sync you control

  • CowAgent★ 47,2 k

    Open-source personal AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, mult…

  • Logseq★ 45,3 k

    Logseq — a privacy-first platform for knowledge management: an outliner over your local Markdown files, with a graph, journals and spaced repetition built in

// plus en productivity

Plus en Productivité

Projets de la catégorie Productivité proches dans le catalogue, les plus proches en étoiles de obsidian-second-brain — continuez à explorer de part et d'autre.

  • OB1★ 4,7 k

    Open Brain (OB1): one database, one AI gateway, one capture channel — shared memory for every AI you use

  • surf★ 3,6 k

    Deta Surf — personal AI notebooks that organise files and webpages and generate notes from them.

  • obsidian-mind★ 5,0 k

    A self-organizing Obsidian vault that gives AI coding agents persistent memory across sessions.

  • llm-for-zotero★ 3,2 k

    An open-source research agent system that lives inside your Zotero library.

  • open-science★ 5,5 k

    A local-first, model-agnostic AI research workbench for reproducible scientific work.

  • vexa★ 2,9 k

    Open-source meeting transcription API for Google Meet, Teams, and Zoom.

Obtenir obsidian-second-brain

Gratuit et ouvert : récupérez le code source, mettez une étoile au dépôt et commencez à créer.