VaultLayer turns a Markdown or Obsidian vault into a rebuildable local database for full-text search, vector retrieval, WikiLinks, metadata, provenance, CLI, and MCP access.
An open-source indexing layer that keeps the vault as plain Markdown files while generating a disposable local database outside the vault and outside the repo. The website page is the public orientation layer; the repository holds implementation details.
Agents should not repeatedly crawl large Markdown folders or treat snippets as context-free truth. They need bounded, cited retrieval with source path, heading, hash, timestamp, excerpt, and enough metadata to explain where an answer came from.
Local vault search, retrieval benchmarks, vector experiments, MCP tool access, and replacing one-off indexing logic in vault-facing apps. It is especially useful when multiple tools need the same rebuildable read model instead of each scanning the vault independently.
The indexer reads Markdown or Obsidian files, builds SQLite/FTS and vector-ready projections, and exposes search results with provenance. Runtime databases, caches, embeddings, and benchmark outputs can be rebuilt from source files instead of becoming a second source of truth.
Private vault text, generated databases, embeddings, caches, and benchmark outputs stay outside the repository. Public examples should use synthetic or explicitly public-safe data, so the project can show the architecture without publishing personal knowledge-base contents.
Pilot-ready local MVP: package smoke tests, doctor checks, benchmark scripts, synthetic messy-vault preflight, and explicit remote-sync boundaries. The immediate proof is not a hosted service; it is a local toolchain that can be rebuilt and verified.
It separates storage from retrieval so viewers, agents, and MCP tools can share one rebuildable vault query layer. That makes provenance and freshness easier to check, while keeping Markdown as the human-editable source.
Open GitHub for the current CLI commands, backend details, benchmarks, and implementation state. This page gives the project shape; the repository carries commands, code, and implementation state.
GitHub remains the place for current files, implementation details, issues, and release state.