Semantic memory for your Obsidian vault
Turns your Obsidian vault into a live, queryable knowledge graph. Connect it once — every agent you use gets semantic search, wikilink traversal, and vault health through standard MCP tools.
How your notes become searchable
A background file watcher fires on every Obsidian save, re-chunks the note, and updates the vector index — no manual step required.
search_vault and receives ranked results with graph boost scores.How search ranks what matters
Semantic similarity is the entry point. Graph proximity and hub importance shape the final ranking.
Before embedding, each chunk is prepended with the note's location, tags, and wikilinks. The vector carries structural context, not just raw text — so similar structure surfaces alongside similar meaning.
Every chunk is stored with folder, tags, wikilinks, and incoming link count as filterable fields. Search can constrain by any dimension before the vector ranking even runs.
After ChromaDB returns candidates, scores are boosted by proximity to a context_note and by hub importance — how many other notes link to this one.
Everything the service can do
Six capabilities built into the same background service.
After every ingest, linker.py appends a ## Related section with wikilinks to semantically similar notes — building the graph without agent intervention.
log_note creates dated notes in typed folders (decision/, lesson/, gotcha/). The watcher picks them up and auto-links them immediately.
vault_status returns orphaned notes, missing frontmatter, broken wikilinks, tag stats, and recent activity — in a single MCP call, no arguments.
move_notes moves one or many files and rewrites every [[wikilink]] across the vault automatically. No orphaned references left behind.
Label propagation over the wikilink graph suggests folder assignments for a single note or the entire vault. Preview before applying, or let the watcher run it automatically.
Standard MCP over stdio. No proprietary SDK, no vendor lock-in. Claude Code, Cursor, Windsurf, Gemini CLI, Zed — if it speaks MCP, it works with archiver-rag.
Structure your vault understands
archiver-rag understands the structure of your vault, not just the text inside it. Every [[link]] is a graph edge that shapes search, clustering, and auto-linking.
Frontmatter is first-class: type, tags, and related are stored as filterable metadata on every chunk.
Tools exposed to any MCP agent
Every capability is a discrete MCP tool — discoverable, composable, and callable from any agent that speaks stdio MCP.
| № | Tool | Function |
|---|---|---|
| 001 | search_vault | Semantic search with graph reranking. Pass context_note to boost wikilink neighbors of a known note. |
| 002 | vault_status | Structure, health, orphaned notes, tag statistics, and recent activity — one call, no arguments. |
| 003 | get_connections | BFS wikilink traversal. Returns direct links or multi-hop neighborhoods up to depth 3. |
| 004 | move_notes | Move one or many files. All [[wikilinks]] across the vault are rewritten automatically after every move. |
| 005 | log_note | Create a dated knowledge note from any agent. The watcher indexes and auto-links it within seconds of creation. |
| 006 | cluster_note | Suggest a folder for one note based on where its wikilink neighbors currently live. Fast, no full-graph pass. |
| 007 | cluster_vault | Label-propagation clustering of the entire vault wikilink graph. Preview with apply=false before committing. |
Up and running in minutes
Install the package, point it at your vault, then register it as an MCP server — done.
pipx install archiver-rag
archiver-rag init
claude mcp add --scope user archiver-rag $(which archiver-rag) serve