Install
$ agentstack add mcp-cyril-36-ai-brain ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
AI Brain
Most AI agents have no shared memory. Claude Desktop forgets what Codex did yesterday. Claude Code starts every session from scattered context. AI Brain makes one Obsidian vault the long-term brain for all of them: every agent reads and writes the same Markdown files through filesystem MCP.
Use it when you want Claude Desktop, Claude Code, Codex, Cowork, and similar tools to coordinate through one durable, human-readable workspace.
What Makes This Different
AGENTS.mdis the master protocol every agent can follow.ACTIVE_CONTEXT.mdis the handoff file every agent reads first.CLAUDE.mdimports the shared protocol for Claude Code.02-Projects/_template/gives each serious project a real workspace: plan, context, TODOs, decisions, logs, prompts, and commands.- Profile notes use
needs-verification: trueso agents do not treat unconfirmed personal facts as ground truth. - Skills and workflows turn repeated agent tasks into reusable playbooks.
The Loop
Capture -> Plan -> Execute -> Reflect -> Learn -> Strategize -> Coordinate.
The vault is designed so agents can capture useful context, plan inside the right project folder, execute with a work log, reflect through reviews, store reusable knowledge, make long-term moves, and coordinate through ACTIVE_CONTEXT.md.
Quick Start
git clone
cp -R ai-brain/template ~/Brain
cd ~/Brain
Then open ~/Brain in Obsidian and configure your AI tools to use the filesystem MCP server for that path.
See [docs/SETUP.md](docs/SETUP.md) for Claude Desktop, Claude Code, Codex, and Cowork setup.
First Prompts To Try
Read AGENTS.md and ACTIVE_CONTEXT.md. Tell me what this vault is set up to do.
Use the project kickoff workflow to create a project folder for my new app.
Run a weekly brain update and create a weekly review note.
Why Not Just...
| Option | Good For | Why This Template Exists | |---|---|---| | A folder of notes | Fast capture | Usually lacks agent rules, handoffs, and project-state conventions. | | PARA | Organizing personal knowledge | This adds AI-agent protocols, verification flags, and repeatable workflows. | | Zettelkasten | Linking ideas over time | This is more operational: projects, decisions, logs, setup, and handoffs. | | Notion or hosted AI memory | Polished databases and hosted UX | Markdown stays portable, local-first, git-friendly, and readable by many tools. |
Repo Layout
ai-brain/
├── docs/
└── template/
├── AGENTS.md
├── CLAUDE.md
├── ACTIVE_CONTEXT.md
├── 01-Profile/
├── 02-Plans/
├── 02-Projects/
├── 03-Skills/
├── 04-Workflows/
├── 05-Knowledge/
├── 06-Journal/
└── 99-Inbox/
The template/ folder is the actual vault. Copy it somewhere else before using it.
Who This Is For
- Developers, students, researchers, and builders using more than one AI coding/chat tool.
- People who want agents to remember project decisions without relying on private chat history.
- Anyone who wants Obsidian as the human-readable UI over an AI-friendly Markdown knowledge base.
What You Get
- Agent protocol:
AGENTS.md - Claude Code shim:
CLAUDE.md - Handoff file:
ACTIVE_CONTEXT.md - Profile placeholders with verification flags
- Project workspace scaffold
- Profile verification, project kickoff, weekly update, and long-term planning skills
- Weekly, monthly, quarterly, and yearly review folders
- Knowledge folders for books, courses, videos, articles, and research papers
License
MIT. Use it, fork it, adapt it, and make it yours.
Source & license
This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Cyril-36
- Source: Cyril-36/ai-brain
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.