mushroomdb

The graph that stays true — and knows who's allowed to see it.

An embedded Rust graph database where edges are a schema declaration: write a rule once, and every write creates, maintains, and retracts the matching edges. Ships a 24-tool MCP server and a live graph of the repository it is pointed at.

Pre-1.0 alpha — APIs and formats may change between minor versions.

Docs · Changelog · Issues

Quick start

Install the Claude Code plugin, from any directory:

claude marketplace add MatthewSherlin/mushroomdb
claude plugin install mushroom@mushroomdb

Open the repository you want graphed and type /mushroom:mushroom. The skill builds the graph on first use and answers with it from then on.

Or install into one project (or your home directory) without the plugin — same skill, invoked bare as /mushroom:

npx mushroomdb install    # /mushroom skill + MCP server + prompt, post-edit and git hooks

Either way, the first thing the assistant does is read the repository back to you. This is a real run against this repository — ingest-git took 2.5 s, map 0.18 s:

mushroomdb map — 431 files, 6,226 symbols, 652 commits, 2 authors · synced 3s ago at 94719fe
clusters (co-change + imports)
  1. <mixed> crates, tests  (86 files, cohesion 0.73)  crates/server/tests/http.rs, algo.rs, crates/server/src/http.rs
  2. <mixed> crates, src  (45 files, cohesion 0.67)  pack.rs, lib.rs, types.rs
  3. ui src, e2e  (26 files, cohesion 0.89)  api.ts, store.ts, classify.ts
  4. crates/code-extract tests, fixtures  (21 files, cohesion 0.99)  lib.rs, extract.rs, mod.rs
  5. ui fonts, public  (18 files, cohesion 0.89)  IBMPlexMono-Medium.woff2, IBMPlexMono-Regular.woff2, IBMPlexSans-Medium.woff2
  6. crates/core-api src, repograph  (17 files, cohesion 0.72)  facts.rs, render.rs, context.rs
  7. <mixed> crates, bindings  (16 files, cohesion 0.99)  crates/core-bench/Cargo.toml, package.json, crates/sim-harness/Cargo.toml
  8. benchmarks adapters, results  (15 files, cohesion 1.00)  run_handrolled.py, datasets.py, handrolled.py
key files (most depended-on)
  crates/code-extract/src/lib.rs 0.05 · crates/server/tests/http.rs 0.04 · crates/code-extract/tests/extract.rs 0.04 · crates/core-api/tests/algo.rs 0.04 · crates/server/src/http.rs 0.03
owners
  Matthew Michael Sherlin 431 files
hot (last 90 days)
  crates/core-api/src/db.rs 175 · README.md 109 · crates/core-rules/src/engine.rs 56 · crates/core-query/src/cypher/exec.rs 54 · crates/cli/src/lib.rs 51
ask me: why does lib.rs co-change with extract.rs? · who owns ui? · what imports http.rs?

From there: context for one file or symbol from every side, impact before an edit, owners, why with the commits that prove a link, recall and remember for durable notes. Full walkthrough: docs/site/code-graph.md.

• Live, not a snapshot. One SET f.top_author_id = … moves the TOP_AUTHOR edge and re-derives that author's KNOWS edges before the write closes — the SET in the GIF above is that one write. An editor hook does the same for your code: touch re-extracts an edited file after every Edit, Write and MultiEdit, in about 180 ms on this repository's graph.
• Retracts instead of going stale. Re-run ingest-git to sync: only new commits replay, deleted files drop their derived edges, and renamed files carry their history to the new path.
• Explains any link. explain names the rule and the score behind an edge, so "which files change together with src/api.rs, and why?" has an answer your assistant can quote instead of a guess.
• Knows who's allowed to see it. Pass a mask with a query and the same graph answers differently per caller; write statements are rejected on masked queries.
• Answers what it said last week. mushroomdb asof ./db --commit 5 --query "…" replays the WAL to a past commit, derived edges included.

Where it fits

What it is

• An embedded, single-binary graph database with a rule engine that maintains edges for you.
• A 24-tool MCP server plus a /mushroom skill and a Claude Code plugin.
• Safe for several processes at once: one writer at a time behind an advisory LOCK file, any number of readers, and every handle picks up a peer's commits by refresh() rather than reopening — so a running serve, an editor hook, a git hook and a CLI command can share one store. docs/site/concurrency.md
• Local-first: your data stays on disk, no cloud service, no LLM in the write path.

What it isn't

• Not a hosted memory service — there is no account, no endpoint, nothing to sign up for.
• Not a vector database. Vector predicates and HNSW are built in; bring your own embeddings.
• Not a Postgres replacement. Single writer, no interactive transactions, memory-first storage.

The differentiator

Most graph databases require you to create edges manually or run a batch similarity script after each load. mushroomdb makes edge creation a schema declaration. A rule like "connect every Person to every Org whose skills list overlaps theirs by at least 50%" is written once:

db.create_rule(RuleDef {
    name: "skill_fit".into(),
    src_label: "Person".into(),
    dst_label: "Org".into(),
    predicate: Predicate::Overlap { field: "skills".into(), min: 0.5 },
    edge_type: "FIT".into(),
    weight_prop: Some("score".into()),
    max_edges: Some(5), // keep the 5 best-matching Orgs per Person (top-k per source)
}).expect("rule");

After that, every insert_node and set_prop evaluates the rule incrementally. The engine writes the edge, stores the Jaccard score, and retracts the edge if the properties later diverge — without any manual work.

Watch it live — a Cypher SET changes one property, the founded_within rule fires, and new scored edges appear in the bundled explorer:

Open the Rules panel, and the Why slide-over shows the exact predicate arithmetic behind every derived edge:

Predicates

Six predicate kinds ship today. They compose via All(...) (AND, score = min) and Any(...) (OR, score = max), nested up to depth 4.

Auto-FK: fields ending in _id whose values match existing node keys get KeyMatch rules created automatically at ingest time. VectorSimilar accepts approximate: true to switch candidate selection to in-tree HNSW (per-query recall min 0.90, mean 0.998 at 5k nodes / dim 1536, fixed-seed probe). Full reference: docs/site/rules.md.

Built on the same engine

• Live subscriptions. subscribe_rule (Rust) and GET /subscribe (WebSocket) stream EdgeFired / EdgeRetracted the moment they hit the WAL — not polled, not batched. Bounded 65,536-event queue; slow consumers get a Lagged { missed: N } marker instead of a disconnect. docs/site/subscriptions.md
• Rule attribution across time. Every derived edge writes a HISTORY-MARKER WAL record carrying the rule name, so edge_history, node_history, and was_linked answer which rule created a link and at which commit. GraphDb::open_at(&dir, 5) replays to a past commit, derived edges included; out-of-range commits return CommitOutOfRange, never wrong data. docs/site/timetravel.md
• Materialized views. Degree counts and neighbor aggregates (sum/avg/min/max) maintained incrementally on every edge change — no cron, no triggers, no stale caches. docs/site/views.md
• Rule suggestions. db.suggest_rules() (or mushroomdb suggest ./db) profiles your data and ranks candidate rules with estimated edge counts and rationale. Seeded sampling, so the same database always returns the same suggestions. No rule is ever applied automatically. docs/site/suggest.md

Agent memory

Graph structure captures the shape of real knowledge — entities, associations, similarity, and lineage — and rule-derived edges keep those associations fresh as new facts arrive.

• Entities map to nodes (Person, Document, Project, Concept, …).
• Associations are edges derived from data: cosine similarity on embeddings, shared field values, FK relationships, geographic proximity. Declare a rule once; every write maintains the matching edges without agent-side bookkeeping.
• Recall has three modes: find_similar by query vector (HNSW when available, brute force otherwise); find_similar by key (neighbors along a rule-derived edge type); query for structured Cypher recall. hybrid_search fuses fulltext and vector results via Reciprocal Rank Fusion.
• Explanations are built in: explain_association shows which rules and scores produced each link, so an agent can cite evidence instead of asserting a conclusion.
• Node masks are the ACL primitive: pass mask: [key1, key2, …] to query to restrict the visible node set. Write statements are rejected on masked queries. docs/site/masks.md
• Schema-as-code: mushroomdb schema apply <dir> <schema.json> idempotently applies rules, views, and fulltext indexes, printing a created/updated/unchanged diff.

Minimal workflow (four tool calls):

upsert_entity  →  create_rule  →  find_similar  →  explain_association
  (store)           (link)           (recall)          (explain)

Eight task tools answer a question about the repository in one call. They are what the skill reaches for, and what tools/list shows first:

The sixteen graph tools reach the store directly. Their descriptions are prefixed Advanced: in tools/list, so an assistant knows which surface is the front door:

Full walkthrough, tool reference, and Claude Desktop setup: docs/site/mcp.md. Skill, plugin, and hook details: docs/site/skill.md.

Install options

claude plugin install mushroom@mushroomdb   # after `claude marketplace add MatthewSherlin/mushroomdb`
npx mushroomdb install            # skill + MCP server + hooks, no toolchain needed
cargo install mushroomdb-cli      # `mushroomdb` binary from crates.io (no embedded UI)
cargo add mushroomdb              # embedded Rust library
pip install mushroomdb            # Python bindings

install writes an MCP entry that runs npx -y mushroomdb@<version>, so the assistant needs nothing installed globally and nothing is copied into your home directory. Point it at a local build with --command <path>. mushroomdb doctor verifies the result end to end — config entry, store, lock, hooks, git hooks, and a real stdio handshake with the configured command.

To see the bundled explorer, write a demo graph and serve it:

mushroomdb demo ./db
mushroomdb serve ./db

Open http://127.0.0.1:8080/. The demo graph has 10 Orgs, 20 Projects, 30 People, and 334 edges — 304 of them derived by seven rule sets. When a token is configured, open http://host:8080/?token=…. Building the binary with the UI embedded, Docker, and the install.sh script are covered in CONTRIBUTING.md.

Role-bound tokens limit a caller to a named subset of nodes. Define roles in schema.json under the roles key (each role has a label selector list), then pass --role-token TOKEN:ROLE (repeatable) when starting the server, or set MUSHROOMDB_ROLE_TOKENS="tok1:role1,tok2:role2". A role token receives only the nodes matching its label selectors — read endpoints return rows filtered to the visible set; write, subscription, and analytics endpoints return 403. Unknown token or role name: 401. The never-widen invariant is enforced in the server: a client-supplied mask is always intersected with the role mask. The MCP interface (mushroomdb mcp) is a stdio JSON-RPC server for local agent use and is not subject to bearer-token or role enforcement.

CLI reference

Concurrency: every CLI write command, the hooks, and a running mushroomdb serve coordinate through one advisory LOCK file in the store directory, so they are safe to run against the same store at the same time. A writer that cannot get the lock within two seconds exits 3 with another mushroomdb process is writing; retry, having written nothing. Readers never take the lock and never wait; recall opens read-only (read_only: true) so an unattended hook can never delay a writer or fail because one is running. What the lock does not give you: cross-process transactions, and subscription events for a peer's writes — a commit absorbed by refresh() is visible on the next read but notifies nobody. Full model: docs/site/concurrency.md.

Full HTTP endpoint reference: docs/site/api.md.

Known limitations

Benchmarks

10,000-node graph (Apple M4 Pro, macOS 15.7.3, arm64), mushroomdb v0.1.1 release build, 2026-08-24. Full methodology and honesty notes: benchmarks/results/head-to-head-10k-v2.md.

Honesty notes:

• mushroomdb numbers are embedded — no network round-trip, no serialization overhead. KùzuDB is also embedded, so its numbers are directly comparable. Neo4j and Memgraph go over bolt/localhost (~0.1–1 ms round-trip per query).
• ★ Two-hop join: same dataset, same warmup policy, all four engines on 5,810,000 INDUSTRY_ALIGNMENT edges. Fresh process → ingest + preload → 3 discarded warmups → median of 10 runs. mushroomdb derives the edges via create_rule; competitors were pre-loaded via UNWIND MERGE or COPY FROM CSV. All engines return 200 rows.
• ★ Earlier v2.1 two-hop values were retracted for cross-engine contamination; the v2 mushroomdb 307 µs figure was retired (measured on a smaller 1M-edge graph). Both are documented in the methodology file rather than quietly dropped.
• ▽ 100k cold-start measured 2026-08-28, warm file cache, cold process, /usr/bin/time -l: V8 snapshot open 0.02 s at 31–41 MiB RSS; snapshot size 1.8 GiB; snapshot write ~35 s. Cold-cache was not measured. See dogfood/results/scale-100k.md.
• Rule engine vs hand-rolled maintenance (10k nodes, 1,000 specialty updates, drift = 0 for all three): per-op expert-written 64.93 min, batched expert-written 24.98 s, rule engine 17.58 s. Both hand-rolled variants were written by the engine team with full knowledge of retraction semantics — drift = 0 is a property of that, not of hand-rolling in general. benchmarks/results/handrolled-vs-rules.md

Architecture

graph-db/
├── crates/
│   ├── core-storage      # Packed adjacency topology + columnar property store + WAL + snapshots
│   ├── core-rules        # linking rules, per-rule indexes, incremental maintenance
│   ├── core-query        # pull-based interpreter; traversal ops + Cypher subset
│   ├── core-api          # the one public Rust interface; typed error enums
│   ├── code-extract      # tree-sitter symbol/import/call extraction; bytes in, facts out
│   ├── arrow-bridge      # results ↔ Arrow buffers
│   ├── server            # axum HTTP + WebSocket; serves UI
│   ├── cli               # mushroomdb binary
│   └── sim-harness       # DST: virtual clock, fault-injecting IO, seeded runner
├── ui/                   # TypeScript + Vite graph explorer
├── bindings/python/      # PyO3 / maturin
└── clients/typescript/   # HTTP + WebSocket client

Dependency rule (inward only): bindings/server/cli → core-api → {core-query, core-rules} → core-storage

Storage uses a dense-id WAL with per-commit fsync (configurable via FsyncPolicy), plus mmap-able V8 rkyv snapshots (12 sections: CSR topology, columnar properties, HNSW blobs, provenance, IVF state, per-node last-change index, and more — zero-copy, no heap allocation on open). V5/V6/V7 stores are auto-migrated to V8 on GraphDb::open. Derived edges are not WAL-logged; they are restored directly from the mmap'd sections. See docs/format-stability.md for the format evolution contract.

Roadmap

Phases 1–4 and Plan 18 all landed. What remains:

Docs

• Quickstart · Rules · Cypher reference · HTTP + MCP API
• The live code graph · Concurrency · Codebase graph
• Install, plugin and hooks · MCP tools
• Time travel · Subscriptions · Views · Rule suggestions
• Masks and access control · Full-text search · Property indexes · Graph algorithms
• Durability and recovery · Panic policy · Testing · Format stability
• Design spec · Moat roadmap · Case study

Building from source, Docker, packaging, and the test gates are in CONTRIBUTING.md.

License

Copyright 2026 Matthew Sherlin.

Dual-licensed under MIT or Apache-2.0, at your option.