Integrations & inspiration

It plugs into what you already use

Any markdown vault in, any MCP-capable agent out. SQLite in the middle. No cloud, no lock-in, no proprietary format.

Supported vaults

Bring your own markdown

The indexer parses YAML frontmatter, [[wiki links]], and heading-path chunks — the common denominator of every markdown PKM tool. No plugin required.

PKM

Obsidian

The reference target. Wiki-links, frontmatter, and Dataview-style inline fields all index cleanly.

PKM

Logseq

Outliner-style markdown with block references. Works as a plain folder of .md files.

PKM

Foam

VS Code-native knowledge base. Same markdown + wiki-link contract the indexer expects.

Anything else that writes plain .md with frontmatter works too — the engine has no vault-app dependency.

Agent harnesses

One vault, many consumers

The MCP protocol means any compliant agent queries the same vault-index.db. bootstrap.sh auto-detects and installs skills into each.

Claude Code

Primary target. Register vdb serve in .mcp.json; skills install to ~/.claude/skills/.

OpenCode

Open-source coding agent. Shares the skills hub via symlink.

Pi

Detected by bootstrap.sh automatically. Shares the same skills hub and MCP server.

Hermes

Another MCP-capable consumer — queries the same vault-index.db through vdb serve.

Cursor

MCP-capable editor — points at the same server, same tools.

Antigravity

Detected by bootstrap alongside the others; per-harness model routing via vdb agents.

Model routing per skill.

Each skill carries a model_tier (1 simple · 2 analysis · 3 deep reasoning). vdb models detects what's available per harness; vdb agents --detect writes harness-specific agent files so cheap tasks run on cheap models.

The protocol

Built on open standards

Protocol

Model Context Protocol

Stdio JSON-RPC 2.0. The MCP server exposes 37 read-mostly vault_* tools any MCP client can call.

Storage

SQLite + FTS5

One file on your disk. FTS5 external-content index over note chunks. No server, no daemon.

Inspiration

Ideas this builds on

Pattern

Karpathy's LLM wiki

Raw notes are immutable source; the DB index is the compiled binary the agent works against. The agent is the programmer, the vault the codebase, the index the build artifact.

Practice

The Obsidian way

Plain files you own, linked by [[wiki links]], readable forever. This project keeps that contract and adds a query layer on top — never replacing the notes.

The hub model

Everything radiates from the vault

Agents don't pull you into their tools — you work from the vault and reach out: planning & delivery, code repositories, infrastructure, and remote services all sync through explicit, user-controlled connections. You own the data. It adapts to your workflow, not the other way around.

planning & delivery code repositories infrastructure remote services notifications
Get started → Star on GitHub