Obsidian Mind plugin
An Obsidian vault template with hooks, commands and skills that gives Claude Code, Codex and Gemini CLI persistent memory across sessions
- 5,528 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Node hooks with an experimental flag run automatically at session start and wrap-up
- The vault accumulates personal notes about people, meetings and decisions
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/obsidian-mindDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Node hooks with an experimental flag run automatically at session start and wrap-up
- The vault accumulates personal notes about people, meetings and decisions
This is third-party code. Review the repository files before installing.
What it does
Obsidian Mind is an Obsidian vault template where hooks written as procedural code handle classification, indexing and note lifecycle, while the agent owns the content: writing notes, linking them together, keeping cards for people and projects. The SessionStart hook loads goals, active projects, recent changes and open tasks at the start of every session. Commands like om-standup, om-dump and om-incident-capture run specific flows: a morning summary, turning a meeting into notes, or reconstructing an incident from a chat link. With the local QMD search tool an MCP server is added for semantic search across the vault, and without it a plain grep through the Obsidian CLI is used instead.
Who it is for. For people running long projects with Claude Code, Codex or Gemini CLI who are tired of re-explaining context every session.
Good fit when
- Work with the agent spans many sessions and context keeps getting lost between them
- You need decisions, meetings and agreements recorded so the agent can find them later on its own
- You already use Obsidian and want the vault to become the agent's memory, not just notes
Not a fit when
- A one-off task in a single session with no need for long-term memory
- Obsidian 1.12+ is not installed and you do not want to set it up for this
Example request
Run /om-standup and remind me where I left off yesterdayLimitations
Full support is claimed only for Claude Code, Codex CLI and Gemini CLI get hooks and commands with no guaranteed feature parity. Semantic search via QMD is optional, but without it the agent falls back to plain grep and is less precise. The hook scripts rely on an experimental Node flag to run TypeScript directly.
How to disable. Delete the vault folder from disk. If installed via ShardMind, you can also remove the .shardmind folder and shard-values.yaml file, the vault keeps working like a plain git clone.
FAQ
Is QMD required?
No, without it the vault works through grep and the Obsidian CLI, but semantic search by meaning and loading topics from brain/ become noticeably sharper with it.
How does ShardMind differ from git clone?
With every value left at its default, a ShardMind install is byte-equivalent to git clone, but it adds a setup wizard tailored to your data and three-way-merge upgrades.
Discussion
No questions or reviews yet.
Related
PluginsPersistent agent memory in an Obsidian vault: saving decisions, search, research and background vault upkeep
An Obsidian vault template for Claude Code: /daily, /weekly, /monthly and /project commands link three-year goals to today's tasks
A build script and templates that turn an Obsidian vault into an AI's working memory: markdown notes, no vector database
An Obsidian vault with skills for work meetings: Gmail transcripts, participants, follow-up tasks and weekly recaps.
Eight Claude Code skills for Obsidian: Granola notes, NotebookLM imports, TaskNotes tasks, recall of past sessions, chat export
An Obsidian vault template where skills sit beside notes, decisions and feedback for Claude Code and Codex.
README in short
The README presents Obsidian Mind as a vault that gives coding agents persistent memory: a session opens with the North Star, active projects and recent changes, and commands like om-standup, om-dump and om-incident-capture turn conversation into structured, linked notes. It explains the architecture: procedural-code hooks handle classification and indexing, the agent handles content. It separately covers QMD for local semantic search through an MCP server running three small offline models. Install is via ShardMind, git clone or a GitHub template, MIT licensed.