How do I install Rms Memory?

Copy the stdio config for Cursor, Claude Desktop, Claude Code, or VS Code on this page. It uses the publisher's install, not a PlugMCP proxy.

Does Rms Memory need an API key?

No required environment variable is listed on this record.

What transport does Rms Memory use?

Rms Memory runs locally over stdio.

When was Rms Memory last checked?

Last checked 2026-10-27. Persistent, local, cross-IDE memory for AI agents — markdown source of truth, LanceDB-powered semantic search, zero cloud dependency

Rms Memory MCP server

🧠 RMS Memory MCP

Version: 1.3.0 (2026-10-13) · companion GUI 1.3.0 (unified numbering)

Persistent, local-first memory for your AI coding agents.

Stop re-explaining your architecture to Cursor, Zed, and Claude Code and other IDEs every single session.

Features • Download • Install • Quick Start • CLI • MCP Tools • Architecture

The Problem

You're developing a single project but switching between different agents — Cursor, Zed, Claude Code, OpenCode, etc. Every one of them loses context of architectural decisions, system requirements, and user preferences the moment you close the tab. You end up re-explaining the same things over and over, or copy-pasting a stale CLAUDE.md between tools.

RMS Memory MCP bridges this gap: a single, isolated, centralized Markdown vault — perfectly structured for LLM consumption — that any MCP-compatible IDE can read from and write to.

🖥️ Prefer a GUI over raw Markdown?

RMS Memory GUI turns your vault into a visual workspace — full graph editor, per-project Git & GitHub sync, one-click Doctor repair, and an AI-assisted Wiki generator (bring your own key). One-time desktop app, works on top of everything below. The MCP server stays 100% free and standalone either way.

✨ Key Features


🗂️ Global Centralized Vaults	Project context lives outside your repo — zero `.mcp` file pollution.
🔍 Hybrid Retrieval (LanceDB)	Embedded Vector Search + Tantivy Full-Text Search for zero-fail context hits.
🌐 Multilingual Semantic Parsing	`fastembed-rs` + `multilingual-e5-small` by default — native Russian & English understanding. Since 1.3.0 `rms-memory embedding use` switches to another bundled model or a local Ollama model (each keeps its own index), and `rms-memory embedding eval` measures hit@k / MRR on your own vault.
🌳 AST Markdown Chunker	`pulldown-cmark`-based chunking keeps code blocks and lists bound to their parent heading.
🧩 Semantic Code Memory	Optional Tree-sitter indexing for Rust, Go, JS/JSX, TS/TSX, Python, C/C++, Java, Ruby, Swift, C#, PHP, Kotlin, SQL, HTML, CSS, and Vue `<script>` blocks; stable segment identities and repeated preambles preserve context when large implementations split. Opt-in `watch` mode reindexes only dirty paths (with full-walk fallback).
🕸️ Knowledge Graph (v1.2.0)	Durable Markdown/code relationships via MCP `rms_graph` (neighbors, path, snapshot, mutations) and optional search `include_graph_neighbors`. Companion GUI still owns the visual GraphView.
🧹 Safe Project Lifecycle	Unregistering preserves vault/index data; permanent GUI deletion requires the exact project key, is confined to the master vault, and never touches source code.
🔀 Federated Corpus Search	Search `vault`, `code`, or `all`; mixed results use Reciprocal Rank Fusion rather than incompatible raw vector distances.
🎯 Bounded Recall (v1.0.7)	`rms_search` returns an inject/abstain envelope with `max_chars`, optional `min_score`, fail-closed errors, and `retrieval_mode` (`hybrid` or short-query `fts_prefer`).
♻️ Knowledge Lifecycle	Frontmatter `status` / `supersedes` / temporal `valid_*` gate Lance recall; soft supersede via `rms_write`; Doctor freshness lint (check 7 of 8); `rms-memory prune` archives aged superseded notes (dry-run default).
📓 Obsidian vaults (v1.3.0)	`rms-memory projects attach-obsidian <PATH>` indexes an existing Obsidian vault in place, strictly read-only (RMS Memory never creates, edits or deletes files in it); `rms-memory import --obsidian <DIR>` copies one into a project vault instead. `[[wikilinks]]`, aliases, headings and embeds resolve the way Obsidian does in the graph, index and doctor.
🌍 Local HTTP transport (v1.3.0)	`rms-memory serve --http` serves MCP Streamable HTTP on `127.0.0.1:7337/mcp` with a bearer token from `~/.rms-memory/http-token`, loopback Host/Origin checks and per-session state — for clients that cannot spawn a stdio process. See SECURITY.md.
🔐 Credential refusal (v1.3.0)	`rms_write` and checkpoint tools refuse text containing well-known token formats (AWS `AKIA…`, GitHub `ghp_…`/`github_pat_…`, OpenAI/Anthropic `sk-…`, Slack, Stripe live, Google, npm, GitLab) or a PEM private key, naming the kind and line but never the value. Doctor check 8 reports such notes already in the vault (report only). Not a guarantee — see SECURITY.md.
⬆️ Update notices (v1.3.0)	`doctor` checks GitHub for a newer release at most once a day (`RMS_MEMORY_NO_UPDATE_CHECK=1` disables it) and warns when the running MCP server binary was replaced on disk; MCP `initialize` repeats the cached notice without network access. Upgrade with the tool that installed it (brew, cargo, package, archive).
🔄 Session Continuity (v1.0.7)	Vault-backed checkpoints (`rms_checkpoint_save/done/load/query`), `rms_overview` project orientation (since 1.3.0 with index lag and git branch / changed files), MCP resources `rms://<project>/…` and prompts (`session_start`, `save_progress`, `record_decision`), `rms_hygiene` consolidation report, `rms_system_instructions` self-bootstrap, editor-agnostic `rms-memory hook` CLI, installer L3 thin adapters, and `pinned` notes that bypass temporal/`min_confidence` recall gates.
🧭 Multi-project MCP routing (v1.0.8+)	Explicit `project` always rebinds the active vault; empty Cursor `roots/list` falls back to process cwd; injected rules require `project: "<key>"` on every memory tool call; `rms-memory inject-rules [--all]` refreshes keys.
🔗 Cross-project federated search (v1.0.9)	Pass `projects: [key, …]` to `rms_search` / `rms_code_search` for read-only RRF federation. Vault/all across multiple projects requires `cross_project_vault=true` on every listed key (hard fail otherwise).
🧊 Concurrent bind cache (v1.0.9)	Up to 4 warm Store+watcher pairs (LRU); multi-root IDE sessions stop thrashing open/close.
🧱 Cargo workspace (v1.0.9+)	`rms-memory-{core,index,vault,cli}` path crates under `crates/`; public umbrella `rms-memory-mcp` keeps stable module paths for the GUI. `cli` hosts cycle-free `gc`/`prune` (1.1.1). See docs/crate-split.md.
📦 Unified Releases	Public assets use `rms_memory_mcp_<version>_<target>.` / `rms_memory_gui_<version>_` on the same `vX.Y.Z` tag (MCP + GUI share numbering). Unversioned names are no longer published.
🪶 Small idle footprint (v1.3.0)	Each `serve` process loads the embedding model only when it needs to embed and releases it after 5 idle minutes (`RMS_MEMORY_MODEL_IDLE_SECS`), so many IDEs with an idle server cost about 50 MB each instead of a model copy each.
⚙️ Dynamic Auto-Installer	`rms-memory install` scans your system and wires itself into every supported IDE.
📜 Rules-as-Code Patching	Non-destructive AST patching of `.cursorrules`, `.zed/assistant.md`, etc. Opt-in by default.
🧪 Durable Vault Writes	`rms_write` creates rolling `.bak` backups and atomically replaces `create`/`replace` targets after fsync, so interrupted writes never expose a truncated Markdown file. Optional `dry_run` previews create/update/noop without touching disk or the index (fingerprint ignores volatile audit stamps).
📜 File git history (v1.1.2)	Derived `code_path` file→commit cache (`rms_file_history` / `rms-memory file-history`); agents should not shell `git log`. Optional search `include_file_history` attaches the last 3 commits per code hit (not with federated `projects`).
📚 Canonical Wiki Isolation	Generated `<vault>/wiki/**` stay Git-synchronized but are excluded from indexes/search/watchers/graph/packs; MCP write and canonical DocumentService also reject wiki paths (wiki-safe writers only).
🛡️ Ten-Point Resiliency	GC, background sync, write-guard snapshots, macOS sandbox bypass, `llms.txt` export, path traversal + injection protection, zombie prevention, graceful shutdown.
🔒 Security Hardened	Panic-free database layer, symlink traversal blocked, JSON-RPC error responses, request size limits. See SECURITY.md and NOTICE.
🧠 Audit Metadata	Every record auto-receives `last_modified_by`, `timestamp`, `confidence`, `source` — agents can filter by reliability.
🔀 Multi-Scope	`--scope` flag supports arbitrary identifiers beyond filesystem paths (thread IDs, lead IDs, etc.).
🖥️ Optional Companion GUI	Paid Tauri desktop app: visual Markdown/graph editor, Git & Vault sync, Doctor dashboard, AI-assisted organizer/Wiki (BYOK), and cross-tool spend tracking — layered on top of the same vault, never required. See GUI-README.md.

📦 Installation

Option 1: Homebrew (macOS Apple Silicon & Linux)

brew tap max-ramas/tap
brew install rms-memory-mcp

Installs a prebuilt binary — no Rust toolchain required. The formula updates automatically with every release.

Not covered by Homebrew: macOS Intel (dropped as of v1.0.1) and Windows (Homebrew doesn't run there — use Option 2 or the .zip below).

Option 2: GitHub release binary

Prebuilt binaries for aarch64-apple-darwin (Apple Silicon), x86_64-unknown-linux-gnu, aarch64-unknown-linux-gnu, and x86_64-pc-windows-msvc are published on every release, along with .deb/.rpm packages for Linux. One-line installers auto-detect your architecture:

curl -fsSL https://raw.githubusercontent.com/max-ramas/rms-memory-mcp/master/scripts/install.sh | bash

irm https://raw.githubusercontent.com/max-ramas/rms-memory-mcp/master/scripts/install.ps1 | iex

Option 3: Build from Source

# 1. Clone the repository
git clone https://github.com/max-ramas/rms-memory-mcp.git
cd rms-memory-mcp

# 2. Build the optimized release binary
cargo build --release

# 3. Add the binary to your global PATH
cp target/release/rms-memory ~/.cargo/bin/

crates.io (`cargo install`)

As of 1.1.0+, tag push publishes only the umbrella crate rms-memory-mcp to crates.io. Internal workspace members (rms-memory-core / index / vault / cli) stay publish = false (path deps for local builds and the companion GUI). Release packaging flattens those crates into a staging tree via scripts/flatten-for-crates-io.py before cargo publish (as of 1.1.1 the staging tree also inlines rms-memory-cli) — see docs/crate-split.md.

cargo install rms-memory-mcp
# Prefer Homebrew or a GitHub release binary if you want a pinned installer.

Optional RMS Memory GUI installers

The companion RMS Memory GUI is a paid, optional Tauri desktop control plane: a visual Markdown/graph editor, per-project and Vault-wide Git/GitHub sync, a Doctor dashboard with one-click repair, an AI-assisted organizer and Wiki generator (bring your own key, proposal-only), and cross-tool spend tracking. The MCP server remains fully standalone: it does not require the GUI, an AI provider, or a GUI license to index, search, sync or serve MCP clients.

See GUI-README.md for the full feature breakdown, supported platforms, installer verification and release-distribution policy.

GUI source is private, but desktop installers are published as binary assets on this repository's GitHub Releases under the matching v<version> tag. Until Apple/Windows signing certificates exist, macOS builds may be unsigned — see GUI-README.md for Gatekeeper notes. The private GUI workflow transfers only the completed .dmg, .msi/.exe, .AppImage, .deb, and .rpm installer files (plus SHA256SUMS.txt when present). With updater signing enabled, the GUI pipeline mirrors signed latest.json, companion .sig files, and macOS *.app.tar.gz onto this public release (URLs rewritten to rms-memory-mcp; in-app Install reads …/releases/latest/download/latest.json). It never mirrors GUI source, build logs, credentials, or private GUI release archives. The same publication flow runs for a v* GUI tag and for a manually dispatched, version-validated GUI release.

🚀 Quick Start

The fastest way to get every IDE on your machine connected:

rms-memory install

This scans ~/.config/ and ~/Library/Application Support/ and hooks rms-memory directly into Cursor, Zed, Claude Code, OpenCode, and others — no manual JSON editing.

Generated Wiki namespace

The optional desktop GUI writes human-readable Wiki pages to <vault>/wiki/. RMS Memory MCP remains AI-free and treats this directory as generated output rather than canonical memory. A shared case-insensitive path policy (src/path_policy.rs, also reused by the GUI) excludes the entire namespace from Markdown/code indexing, vector and full-text retrieval, watchers, the durable graph and Wiki context packs. Write isolation matches that policy: rms_write requires .md and rejects wiki/**; canonical DocumentService list/read/write APIs exclude or reject wiki; Wiki page mutations use wiki-safe methods that skip memory audit-frontmatter injection. Linked-document link: resolution always re-checks that the canonical target stays inside the vault. Full or incremental sync removes legacy Wiki-derived records by path without deleting the files, and doctor reports the isolation state explicitly.

For virtual projects without a filesystem path (threads, leads, etc.), use --scope:

rms-memory --scope "thread:abc-123" serve

Use multiple isolated scopes

A scope is an isolation boundary for a vault and its index. Without --scope, RMS Memory uses the canonical current working directory; an explicit filesystem path addresses that same kind of project vault. Any other non-empty identifier creates an isolated virtual vault:

rms-memory serve                                      # current project scope
rms-memory --scope "/home/user/my-project" serve     # explicit project scope
rms-memory --scope "thread:abc-123" serve             # virtual thread scope
rms-memory --scope "product:acme" serve                # virtual product scope

For project knowledge plus per-thread history, query each scope explicitly and merge the results in the caller. RMS Memory intentionally does not mix scopes implicitly. Scope IDs may not be empty or exceed 512 characters; absolute and .//../ values are resolved as paths, while all other values are opaque identifiers.

When using min_confidence, start with an unfiltered search. Use 0.3–0.5 for broad refinement and reserve 0.7+ for verified canonical facts; records without a confidence value remain visible.

Configure your vault

The simplest way to configure the server is to run the interactive setup wizard. You don't need to memorize any CLI flags — just run:

rms-memory config

(Alternatively, set the vault root directly with rms-memory config --vault-path ~/MyVaults/, then run rms-memory init in each repository you want to register.)

cd /path/to/project
rms-memory init

This creates the project mapping in ~/.rms-memory/registry.toml and provisions its isolated, structured vault. Routine MCP discovery is read-only and fail-closed: it never creates a project from /, never falls back to a shared global vault, and never guesses between multiple registered projects.

~/MyVaults/
  └── <ProjectKey>/
      ├── rules/
      ├── decisions/
      ├── architecture/
      ├── artifacts/
      ├── docs/
      └── api/

Optional semantic code memory

Markdown memory remains the default corpus. Semantic source indexing is separate, supports all bundled language adapters, and never changes source files:

rms-memory reindex --code  # build/update only derived code memory
rms-memory reindex --all   # refresh Markdown vault + code memory

Registered projects support code_index_mode = "off" | "manual" | "watch"; the default is off. Set it from the project root with rms-memory config --code-index-mode watch (or add --scope <project-path>). watch is explicitly opt-in, coalesces supported source saves for three seconds, and reindexes only the dirty paths (try_index_code_paths) with a full-walk fallback when the index is cold, the dirty set is empty/oversized (>200), or the watcher channel overflows. Concurrent IDE processes share a completion marker so an unchanged workspace stays idle. Code search results include their source language.

Perf smoke for large fixtures: ./scripts/bench_large_vault.sh [notes] [code_files].

Language selection is project-scoped and defaults to every bundled adapter:

rms-memory config --code-languages auto
rms-memory config --code-languages go,typescript,tsx,vue

Supported names are rust, go, javascript, jsx, typescript, tsx, python, c, cpp, java, ruby, swift, vue, csharp, php, kotlin, sql, html, and css. Generated paths (node_modules, .next, .nuxt, target, vendor, and coverage) are always excluded. Ambiguous .h files are indexed as C exactly once; use .hpp, .hh, or .hxx for C++ headers. Vue indexes only inline JavaScript/TypeScript <script> contents and maps results back to the .vue host file; templates, styles, script setup macros, and external src scripts remain outside v1.0.5 semantic extraction. HTML files contribute inline JavaScript <script> blocks (mapped back to the .html file) and elements with an id; CSS files contribute rule sets by selector and @keyframes. SQL files contribute CREATE TABLE / VIEW / TYPE / FUNCTION / PROCEDURE / TRIGGER statements.

Use an existing Obsidian vault

rms-memory projects attach-obsidian ~/Obsidian/Notes --key notes   # in place, read-only
rms-memory import --obsidian ~/Obsidian/Notes --dry-run             # or copy into this project's vault

An attached vault is indexed and searchable (pass project: "notes"), but RMS Memory never creates, edits, moves or deletes files in it: rms_write, checkpoints, prune and hooks refuse, and projects remove only drops the registration. The index lives under ~/.rms-memory/dbs/ like any other. Obsidian links ([[Note]], [[Note|label]], [[Note#Heading]], ![[image.png]], frontmatter aliases) resolve in the graph, the index and doctor.

Embedding model and retrieval eval

The default model is multilingual-e5-small (384 dimensions, downloaded once). To compare it with others on your own notes and switch:

rms-memory embedding list
rms-memory embedding eval --models multilingual-e5-small,bge-m3 --show-misses
rms-memory embedding use bge-m3              # or: ollama:nomic-embed-text (local Ollama)

eval turns each note's first heading into a query (or reads --queries file.jsonl with {"query": "...", "expect": ["path.md"]} lines) and reports hit@1, hit@k and MRR for embeddings alone and for hybrid search. Each model keeps its own index under the project's dbs/ folder, so switching back reuses the old one. After use, restart running MCP servers (IDE, GUI); the next sync builds the new index. Ollama URLs must be loopback (http://127.0.0.1:11434 by default).

🛠 CLI Commands

Command	Description
`rms-memory serve [--http] [--port 7337]`	Starts the JSON-RPC stdio server (auto-triggered by your IDE). `--http` serves MCP Streamable HTTP on `127.0.0.1` instead, with a bearer token.
`rms-memory init`	Registers a project into the global registry. `--dry-run` supported. `--full` forces creation of all IDE rule templates.
`rms-memory inject-rules [--all]`	Re-injects managed IDE rule blocks with the concrete registry `project` key (existing files only, unless `--full`). Use after template updates. Fail-closed: a single unregistered path is refused rather than injected with a guessed key — run `init` there first, or use `--all` to refresh every registered project.
`rms-memory import`	Scans for existing docs (`README.md`, `docs/`, `ADR/`) and imports them — interactively or via `--auto-import`. `--obsidian <DIR> [--into obsidian] [--dry-run]` copies an Obsidian vault (notes, attachments, folders) into the project vault without touching the source or overwriting files.
`rms-memory install`	Hooks the server into supported IDEs. `--dry-run` supported.
`rms-memory uninstall`	Removes the server from all discovered IDE configurations.
`rms-memory doctor`	Runs 8-point vault health diagnostics (the 8th reports credentials stored in notes, never changing them) and prints update / restart notices. `--repair-frontmatter` safely repairs duplicate, missing, and known attached frontmatter IDs with backups; arbitrary invalid YAML is reported but never rewritten automatically.
`rms-memory config`	Without flags: prints global + current-project settings, then offers interactive global editing. Any flag runs non-interactively. Global: `--vault-path`, `--auto-add`, `--inject-rules`, `--auto-import skip\|link\|import_organize\|import`, `--max-backups N`. Project (cwd or `--scope <path>`): `--code-index-mode off\|manual\|watch`, `--code-languages auto\|<comma-list>`, `--include <globs>`, `--exclude <globs>`, `--cross-project-vault true\|false`.
`rms-memory reindex [--vault\|--code\|--all]`	Refreshes Markdown memory (default), derived semantic code memory, or both.
`rms-memory sync`	Incremental LanceDB delete-then-insert sync (also runs automatically during `serve`).
`rms-memory gc`	Prunes orphaned LanceDB indices belonging to deleted vaults.
`rms-memory prune [--older-than-days N] [--apply]`	Archives superseded notes older than N days (default 30) under `artifacts/pruned/YYYY-MM-DD/`. Dry-run by default; never deletes. Distinct from `gc` (orphan DBs) and from supersession (lifecycle marking).
`rms-memory file-history catch-up\|reindex\|query`	Derived `code_path` git file→commit cache (Lance). `reindex` requires `--project`. Prefer MCP `rms_file_history` for agents.
`rms-memory features`	Live GUI/AI status banner + capability catalog (`GUI`/`AI` tags; soft yellow when GUI absent, gray when installed). Informational only — MCP core is never paywalled.
`rms-memory graph status\|ensure\|neighbors\|path\|snapshot\|export-dot\|…`	Durable knowledge graph (same actions as MCP `rms_graph`). Mutations require `--project`.
`rms-memory log`	Tails the telemetry log (`~/.rms-memory/rms.log`; moved to `rms.log.1` when a process starts and it exceeds 20 MB).
`rms-memory export-llms`	Compiles the current vault into a single `llms.txt` payload.
`rms-memory projects list`	Lists registered project keys and their code/vault paths.
`rms-memory projects locate --project <key>`	Resolves one registered project key.
`rms-memory projects resolve-key --path <dir>`	Looks up the registry key for a code path (including post-migrate redirects).
`rms-memory projects migrate --project <key> --to <new-path>`	Moves/renames a registered project after the repo folder changed. Plans key rename, vault/db moves, and `link:` repairs; supports `--dry-run`, `--no-repair-links`, `--strict-git`. Prefer this over recreating `.git` or re-running `init`.
`rms-memory projects remove <key>`	Removes an erroneous project registration while preserving its vault files.
`rms-memory projects attach-obsidian <PATH> [--key KEY]`	Registers an existing Obsidian vault in place, read-only: indexed and searchable, never modified. The read-only flag is kept in `~/.rms-memory` (`registry.toml` and `read-only-vaults`); if that directory is deleted or recreated, attach the vault again.
`rms-memory embedding list\|use <id>\|eval`	Lists the bundled embedding models, switches the model (`ollama:<model>` for a local Ollama), or scores retrieval quality (hit@1, hit@k, MRR) on the project's vault, for embeddings alone and hybrid search. Restart MCP servers after `use`; the next sync builds the model's index.
`rms-memory hook --event <e>`	Editor-agnostic continuity hook (`session_start`, `pre_compact`, `session_stop`); JSON on stdout. `--project <key>` or unique cwd resolution (fail-closed); `--apply` creates/updates or closes a checkpoint.
All commands	Accept `--scope <id>` to target arbitrary isolated vaults (threads, leads, etc.).

🔌 MCP Tools Exposed

Tool descriptions are written to be action-oriented, so agents use the vault proactively without being asked.

MCP resources (1.3.0): rms://<project>/overview, rms://<project>/instructions, rms://<project>/checkpoint/<name> and rms://<project>/note/<path> are read-only views for clients that attach context by URI. MCP prompts: session_start, save_progress (name) and record_decision (topic).

The server resolves an explicit scope or legacy rootUri, then negotiates MCP roots/list. If a client exposes neither (or opens several registered roots), pass the short registry key in project; injected agent rules contain the correct key for that repository. rms_projects lists valid keys without requiring a bound workspace. An explicit project on any tool call always wins and rebinds the active vault — one long-lived MCP process can serve every registered project. Without project, ambiguity stays fail-closed (no silent pick-first).

To remove an accidental registration without deleting its Markdown vault:

rms-memory projects remove <key>

If the repository folder was moved or renamed, do not recreate .git or re-run init from scratch. Plan and apply a migrate instead:

rms-memory projects migrate --project <key> --to /new/path/to/repo --dry-run
rms-memory projects migrate --project <key> --to /new/path/to/repo
rms-memory projects resolve-key --path /new/path/to/repo

The CLI command projects remove is intentionally non-destructive. The companion GUI exposes a separate Delete project and data action for permanent cleanup of the registration, Markdown vault, and derived index. It requires typing the exact project key and accepts only a dedicated child of the configured master vault; the repository source path is explicitly excluded from deletion.

🏗 Architecture Highlights

Implementation lives in path-only crates under crates/ (rms-memory-core, rms-memory-index, rms-memory-vault, rms-memory-cli for cycle-free gc/prune). The published product remains the root umbrella rms-memory-mcp (binary + MCP server + tools + rules injector + serve), which re-exports every former rms_memory_mcp::<module> path so the companion GUI keeps stable imports. Heavy deps (lancedb, ort, fastembed, tree-sitter) concentrate in rms-memory-index. Details: docs/crate-split.md.

A central ~/.rms-memory/registry.toml routes every project to an isolated vault, computed from a hash of the project path. No .mcp files, no per-repo config — global MCP entries (e.g. Zed's settings.json) can target any workspace automatically.

The transport-neutral ProjectService is the single implementation used by the CLI and companion GUI. Registry mutation remains revisioned through ConfigManager; deletion validates canonical paths before unregistering and returns structured warnings if filesystem cleanup cannot be completed.

Instead of duplicating existing docs into the vault, rms-memory import can create lightweight Link Files — Markdown stubs with a link: <path> frontmatter property. Reads/writes are transparently redirected to the source file, while the vector index still respects the vault's directory structure.

Embedded LanceDB (~/.rms-memory/dbs/) combines vector similarity with full-text search, so a query never comes back empty just because the exact keywords didn't match.

Human-authored Markdown and derived Rust code live in separate tables. Code chunks carry stable symbol identities, line ranges, and preambles; unchanged chunks reuse their vectors. corpus=all fuses independently ranked result sets with Reciprocal Rank Fusion, avoiding any assumption that distances from the two corpora are comparable.

Graph nodes and edges are deliberately independent of retrieval chunk boundaries. Markdown links, Rust imports, trait implementations, and lexical call hints can be reconciled as derived relationships; user-created edges and suppress/restore overrides persist across reindexing. Current Rust call edges are syntax-level hints, not a compiler-accurate call graph.

pulldown-cmark parses the Markdown AST directly. Chunks are built by walking up to the parent heading, with a strict 1500-character boundary and ~200-character overlapping window for oversized code blocks — no mid-sentence truncation.

Path traversal + filter injection prevention
Zombie process prevention (watcher shutdown on EOF + std::process::exit(0))
Graceful shutdown (SIGINT/Ctrl+C handler)
macOS sandbox bypass for fastembed model downloads
rms-memory gc — orphaned vector store pruning
PID-aware per-project writer lock and read-only background synchronization across IDE processes
Markdown watcher plus an explicitly opt-in, 3s-debounced code watcher with path-scoped reindex and shared-generation suppression
Write-guard snapshotting with rolling .bak backups (default: 5)
Isolated telemetry logging (~/.rms-memory/rms.log)
llms.txt export for flat, decoupled LLM ingestion

Live MCP requests have been verified for rms_search(corpus=vault|code|all) and rms_code_search. On this repository, reindex --code indexed 43 Rust files into 298 semantic items and 438 segments with all vectors reused on an unchanged run. An isolated five-server watcher run coalesced rapid saves into one shared completion-marker update; a later real-project stress gate completed concurrent GeoMail, License Server, RMS Monitoring, and GeoTax Site indexing, then seven MCP servers (after four IDE restarts) stayed at 0.0% CPU with no background reindex.

🧩 Supported IDEs

IDE	Auto-Install	Rules Injection
Cursor	✅	`.cursorrules`
Zed	✅	`.zed/assistant.md`
Claude Code	✅	`.claude/CLAUDE.md`
OpenCode	✅	—
Codex	✅	—
VS Code	✅	—
Antigravity	✅	—

📄 License

MIT License — see LICENSE for details.

🧠 RMS Memory MCP

Version: 1.3.0 (2026-10-13) · companion GUI 1.3.0 (unified numbering)

Persistent, local-first memory for your AI coding agents.

Stop re-explaining your architecture to Cursor, Zed, and Claude Code and other IDEs every single session.

Features • Download • Install • Quick Start • CLI • MCP Tools • Architecture

The Problem

RMS Memory MCP bridges this gap: a single, isolated, centralized Markdown vault — perfectly structured for LLM consumption — that any MCP-compatible IDE can read from and write to.

🖥️ Prefer a GUI over raw Markdown?

✨ Key Features


🗂️ Global Centralized Vaults	Project context lives outside your repo — zero `.mcp` file pollution.
🔍 Hybrid Retrieval (LanceDB)	Embedded Vector Search + Tantivy Full-Text Search for zero-fail context hits.
🌐 Multilingual Semantic Parsing	`fastembed-rs` + `multilingual-e5-small` by default — native Russian & English understanding. Since 1.3.0 `rms-memory embedding use` switches to another bundled model or a local Ollama model (each keeps its own index), and `rms-memory embedding eval` measures hit@k / MRR on your own vault.
🌳 AST Markdown Chunker	`pulldown-cmark`-based chunking keeps code blocks and lists bound to their parent heading.
🧩 Semantic Code Memory	Optional Tree-sitter indexing for Rust, Go, JS/JSX, TS/TSX, Python, C/C++, Java, Ruby, Swift, C#, PHP, Kotlin, SQL, HTML, CSS, and Vue `<script>` blocks; stable segment identities and repeated preambles preserve context when large implementations split. Opt-in `watch` mode reindexes only dirty paths (with full-walk fallback).
🕸️ Knowledge Graph (v1.2.0)	Durable Markdown/code relationships via MCP `rms_graph` (neighbors, path, snapshot, mutations) and optional search `include_graph_neighbors`. Companion GUI still owns the visual GraphView.
🧹 Safe Project Lifecycle	Unregistering preserves vault/index data; permanent GUI deletion requires the exact project key, is confined to the master vault, and never touches source code.
🔀 Federated Corpus Search	Search `vault`, `code`, or `all`; mixed results use Reciprocal Rank Fusion rather than incompatible raw vector distances.
🎯 Bounded Recall (v1.0.7)	`rms_search` returns an inject/abstain envelope with `max_chars`, optional `min_score`, fail-closed errors, and `retrieval_mode` (`hybrid` or short-query `fts_prefer`).
♻️ Knowledge Lifecycle	Frontmatter `status` / `supersedes` / temporal `valid_*` gate Lance recall; soft supersede via `rms_write`; Doctor freshness lint (check 7 of 8); `rms-memory prune` archives aged superseded notes (dry-run default).
📓 Obsidian vaults (v1.3.0)	`rms-memory projects attach-obsidian <PATH>` indexes an existing Obsidian vault in place, strictly read-only (RMS Memory never creates, edits or deletes files in it); `rms-memory import --obsidian <DIR>` copies one into a project vault instead. `[[wikilinks]]`, aliases, headings and embeds resolve the way Obsidian does in the graph, index and doctor.
🌍 Local HTTP transport (v1.3.0)	`rms-memory serve --http` serves MCP Streamable HTTP on `127.0.0.1:7337/mcp` with a bearer token from `~/.rms-memory/http-token`, loopback Host/Origin checks and per-session state — for clients that cannot spawn a stdio process. See SECURITY.md.
🔐 Credential refusal (v1.3.0)	`rms_write` and checkpoint tools refuse text containing well-known token formats (AWS `AKIA…`, GitHub `ghp_…`/`github_pat_…`, OpenAI/Anthropic `sk-…`, Slack, Stripe live, Google, npm, GitLab) or a PEM private key, naming the kind and line but never the value. Doctor check 8 reports such notes already in the vault (report only). Not a guarantee — see SECURITY.md.
⬆️ Update notices (v1.3.0)	`doctor` checks GitHub for a newer release at most once a day (`RMS_MEMORY_NO_UPDATE_CHECK=1` disables it) and warns when the running MCP server binary was replaced on disk; MCP `initialize` repeats the cached notice without network access. Upgrade with the tool that installed it (brew, cargo, package, archive).
🔄 Session Continuity (v1.0.7)	Vault-backed checkpoints (`rms_checkpoint_save/done/load/query`), `rms_overview` project orientation (since 1.3.0 with index lag and git branch / changed files), MCP resources `rms://<project>/…` and prompts (`session_start`, `save_progress`, `record_decision`), `rms_hygiene` consolidation report, `rms_system_instructions` self-bootstrap, editor-agnostic `rms-memory hook` CLI, installer L3 thin adapters, and `pinned` notes that bypass temporal/`min_confidence` recall gates.
🧭 Multi-project MCP routing (v1.0.8+)	Explicit `project` always rebinds the active vault; empty Cursor `roots/list` falls back to process cwd; injected rules require `project: "<key>"` on every memory tool call; `rms-memory inject-rules [--all]` refreshes keys.
🔗 Cross-project federated search (v1.0.9)	Pass `projects: [key, …]` to `rms_search` / `rms_code_search` for read-only RRF federation. Vault/all across multiple projects requires `cross_project_vault=true` on every listed key (hard fail otherwise).
🧊 Concurrent bind cache (v1.0.9)	Up to 4 warm Store+watcher pairs (LRU); multi-root IDE sessions stop thrashing open/close.
🧱 Cargo workspace (v1.0.9+)	`rms-memory-{core,index,vault,cli}` path crates under `crates/`; public umbrella `rms-memory-mcp` keeps stable module paths for the GUI. `cli` hosts cycle-free `gc`/`prune` (1.1.1). See docs/crate-split.md.
📦 Unified Releases	Public assets use `rms_memory_mcp_<version>_<target>.` / `rms_memory_gui_<version>_` on the same `vX.Y.Z` tag (MCP + GUI share numbering). Unversioned names are no longer published.
🪶 Small idle footprint (v1.3.0)	Each `serve` process loads the embedding model only when it needs to embed and releases it after 5 idle minutes (`RMS_MEMORY_MODEL_IDLE_SECS`), so many IDEs with an idle server cost about 50 MB each instead of a model copy each.
⚙️ Dynamic Auto-Installer	`rms-memory install` scans your system and wires itself into every supported IDE.
📜 Rules-as-Code Patching	Non-destructive AST patching of `.cursorrules`, `.zed/assistant.md`, etc. Opt-in by default.
🧪 Durable Vault Writes	`rms_write` creates rolling `.bak` backups and atomically replaces `create`/`replace` targets after fsync, so interrupted writes never expose a truncated Markdown file. Optional `dry_run` previews create/update/noop without touching disk or the index (fingerprint ignores volatile audit stamps).
📜 File git history (v1.1.2)	Derived `code_path` file→commit cache (`rms_file_history` / `rms-memory file-history`); agents should not shell `git log`. Optional search `include_file_history` attaches the last 3 commits per code hit (not with federated `projects`).
📚 Canonical Wiki Isolation	Generated `<vault>/wiki/**` stay Git-synchronized but are excluded from indexes/search/watchers/graph/packs; MCP write and canonical DocumentService also reject wiki paths (wiki-safe writers only).
🛡️ Ten-Point Resiliency	GC, background sync, write-guard snapshots, macOS sandbox bypass, `llms.txt` export, path traversal + injection protection, zombie prevention, graceful shutdown.
🔒 Security Hardened	Panic-free database layer, symlink traversal blocked, JSON-RPC error responses, request size limits. See SECURITY.md and NOTICE.
🧠 Audit Metadata	Every record auto-receives `last_modified_by`, `timestamp`, `confidence`, `source` — agents can filter by reliability.
🔀 Multi-Scope	`--scope` flag supports arbitrary identifiers beyond filesystem paths (thread IDs, lead IDs, etc.).
🖥️ Optional Companion GUI	Paid Tauri desktop app: visual Markdown/graph editor, Git & Vault sync, Doctor dashboard, AI-assisted organizer/Wiki (BYOK), and cross-tool spend tracking — layered on top of the same vault, never required. See GUI-README.md.

📦 Installation

Option 1: Homebrew (macOS Apple Silicon & Linux)

brew tap max-ramas/tap
brew install rms-memory-mcp

Installs a prebuilt binary — no Rust toolchain required. The formula updates automatically with every release.

Not covered by Homebrew: macOS Intel (dropped as of v1.0.1) and Windows (Homebrew doesn't run there — use Option 2 or the .zip below).

Option 2: GitHub release binary

curl -fsSL https://raw.githubusercontent.com/max-ramas/rms-memory-mcp/master/scripts/install.sh | bash

irm https://raw.githubusercontent.com/max-ramas/rms-memory-mcp/master/scripts/install.ps1 | iex

Option 3: Build from Source

# 1. Clone the repository
git clone https://github.com/max-ramas/rms-memory-mcp.git
cd rms-memory-mcp

# 2. Build the optimized release binary
cargo build --release

# 3. Add the binary to your global PATH
cp target/release/rms-memory ~/.cargo/bin/

crates.io (`cargo install`)

cargo install rms-memory-mcp
# Prefer Homebrew or a GitHub release binary if you want a pinned installer.

Optional RMS Memory GUI installers

See GUI-README.md for the full feature breakdown, supported platforms, installer verification and release-distribution policy.

🚀 Quick Start

The fastest way to get every IDE on your machine connected:

rms-memory install

This scans ~/.config/ and ~/Library/Application Support/ and hooks rms-memory directly into Cursor, Zed, Claude Code, OpenCode, and others — no manual JSON editing.

Generated Wiki namespace

For virtual projects without a filesystem path (threads, leads, etc.), use --scope:

rms-memory --scope "thread:abc-123" serve

Use multiple isolated scopes

rms-memory serve                                      # current project scope
rms-memory --scope "/home/user/my-project" serve     # explicit project scope
rms-memory --scope "thread:abc-123" serve             # virtual thread scope
rms-memory --scope "product:acme" serve                # virtual product scope

When using min_confidence, start with an unfiltered search. Use 0.3–0.5 for broad refinement and reserve 0.7+ for verified canonical facts; records without a confidence value remain visible.

Configure your vault

The simplest way to configure the server is to run the interactive setup wizard. You don't need to memorize any CLI flags — just run:

rms-memory config

(Alternatively, set the vault root directly with rms-memory config --vault-path ~/MyVaults/, then run rms-memory init in each repository you want to register.)

cd /path/to/project
rms-memory init

~/MyVaults/
  └── <ProjectKey>/
      ├── rules/
      ├── decisions/
      ├── architecture/
      ├── artifacts/
      ├── docs/
      └── api/

Optional semantic code memory

Markdown memory remains the default corpus. Semantic source indexing is separate, supports all bundled language adapters, and never changes source files:

rms-memory reindex --code  # build/update only derived code memory
rms-memory reindex --all   # refresh Markdown vault + code memory

Perf smoke for large fixtures: ./scripts/bench_large_vault.sh [notes] [code_files].

Language selection is project-scoped and defaults to every bundled adapter:

rms-memory config --code-languages auto
rms-memory config --code-languages go,typescript,tsx,vue

Use an existing Obsidian vault

rms-memory projects attach-obsidian ~/Obsidian/Notes --key notes   # in place, read-only
rms-memory import --obsidian ~/Obsidian/Notes --dry-run             # or copy into this project's vault

Embedding model and retrieval eval

The default model is multilingual-e5-small (384 dimensions, downloaded once). To compare it with others on your own notes and switch:

rms-memory embedding list
rms-memory embedding eval --models multilingual-e5-small,bge-m3 --show-misses
rms-memory embedding use bge-m3              # or: ollama:nomic-embed-text (local Ollama)

🛠 CLI Commands

Command	Description
`rms-memory serve [--http] [--port 7337]`	Starts the JSON-RPC stdio server (auto-triggered by your IDE). `--http` serves MCP Streamable HTTP on `127.0.0.1` instead, with a bearer token.
`rms-memory init`	Registers a project into the global registry. `--dry-run` supported. `--full` forces creation of all IDE rule templates.
`rms-memory inject-rules [--all]`	Re-injects managed IDE rule blocks with the concrete registry `project` key (existing files only, unless `--full`). Use after template updates. Fail-closed: a single unregistered path is refused rather than injected with a guessed key — run `init` there first, or use `--all` to refresh every registered project.
`rms-memory import`	Scans for existing docs (`README.md`, `docs/`, `ADR/`) and imports them — interactively or via `--auto-import`. `--obsidian <DIR> [--into obsidian] [--dry-run]` copies an Obsidian vault (notes, attachments, folders) into the project vault without touching the source or overwriting files.
`rms-memory install`	Hooks the server into supported IDEs. `--dry-run` supported.
`rms-memory uninstall`	Removes the server from all discovered IDE configurations.
`rms-memory doctor`	Runs 8-point vault health diagnostics (the 8th reports credentials stored in notes, never changing them) and prints update / restart notices. `--repair-frontmatter` safely repairs duplicate, missing, and known attached frontmatter IDs with backups; arbitrary invalid YAML is reported but never rewritten automatically.
`rms-memory config`	Without flags: prints global + current-project settings, then offers interactive global editing. Any flag runs non-interactively. Global: `--vault-path`, `--auto-add`, `--inject-rules`, `--auto-import skip\|link\|import_organize\|import`, `--max-backups N`. Project (cwd or `--scope <path>`): `--code-index-mode off\|manual\|watch`, `--code-languages auto\|<comma-list>`, `--include <globs>`, `--exclude <globs>`, `--cross-project-vault true\|false`.
`rms-memory reindex [--vault\|--code\|--all]`	Refreshes Markdown memory (default), derived semantic code memory, or both.
`rms-memory sync`	Incremental LanceDB delete-then-insert sync (also runs automatically during `serve`).
`rms-memory gc`	Prunes orphaned LanceDB indices belonging to deleted vaults.
`rms-memory prune [--older-than-days N] [--apply]`	Archives superseded notes older than N days (default 30) under `artifacts/pruned/YYYY-MM-DD/`. Dry-run by default; never deletes. Distinct from `gc` (orphan DBs) and from supersession (lifecycle marking).
`rms-memory file-history catch-up\|reindex\|query`	Derived `code_path` git file→commit cache (Lance). `reindex` requires `--project`. Prefer MCP `rms_file_history` for agents.
`rms-memory features`	Live GUI/AI status banner + capability catalog (`GUI`/`AI` tags; soft yellow when GUI absent, gray when installed). Informational only — MCP core is never paywalled.
`rms-memory graph status\|ensure\|neighbors\|path\|snapshot\|export-dot\|…`	Durable knowledge graph (same actions as MCP `rms_graph`). Mutations require `--project`.
`rms-memory log`	Tails the telemetry log (`~/.rms-memory/rms.log`; moved to `rms.log.1` when a process starts and it exceeds 20 MB).
`rms-memory export-llms`	Compiles the current vault into a single `llms.txt` payload.
`rms-memory projects list`	Lists registered project keys and their code/vault paths.
`rms-memory projects locate --project <key>`	Resolves one registered project key.
`rms-memory projects resolve-key --path <dir>`	Looks up the registry key for a code path (including post-migrate redirects).
`rms-memory projects migrate --project <key> --to <new-path>`	Moves/renames a registered project after the repo folder changed. Plans key rename, vault/db moves, and `link:` repairs; supports `--dry-run`, `--no-repair-links`, `--strict-git`. Prefer this over recreating `.git` or re-running `init`.
`rms-memory projects remove <key>`	Removes an erroneous project registration while preserving its vault files.
`rms-memory projects attach-obsidian <PATH> [--key KEY]`	Registers an existing Obsidian vault in place, read-only: indexed and searchable, never modified. The read-only flag is kept in `~/.rms-memory` (`registry.toml` and `read-only-vaults`); if that directory is deleted or recreated, attach the vault again.
`rms-memory embedding list\|use <id>\|eval`	Lists the bundled embedding models, switches the model (`ollama:<model>` for a local Ollama), or scores retrieval quality (hit@1, hit@k, MRR) on the project's vault, for embeddings alone and hybrid search. Restart MCP servers after `use`; the next sync builds the model's index.
`rms-memory hook --event <e>`	Editor-agnostic continuity hook (`session_start`, `pre_compact`, `session_stop`); JSON on stdout. `--project <key>` or unique cwd resolution (fail-closed); `--apply` creates/updates or closes a checkpoint.
All commands	Accept `--scope <id>` to target arbitrary isolated vaults (threads, leads, etc.).

🔌 MCP Tools Exposed

Tool descriptions are written to be action-oriented, so agents use the vault proactively without being asked.

To remove an accidental registration without deleting its Markdown vault:

rms-memory projects remove <key>

If the repository folder was moved or renamed, do not recreate .git or re-run init from scratch. Plan and apply a migrate instead:

rms-memory projects migrate --project <key> --to /new/path/to/repo --dry-run
rms-memory projects migrate --project <key> --to /new/path/to/repo
rms-memory projects resolve-key --path /new/path/to/repo

🏗 Architecture Highlights

Embedded LanceDB (~/.rms-memory/dbs/) combines vector similarity with full-text search, so a query never comes back empty just because the exact keywords didn't match.

Path traversal + filter injection prevention
Zombie process prevention (watcher shutdown on EOF + std::process::exit(0))
Graceful shutdown (SIGINT/Ctrl+C handler)
macOS sandbox bypass for fastembed model downloads
rms-memory gc — orphaned vector store pruning
PID-aware per-project writer lock and read-only background synchronization across IDE processes
Markdown watcher plus an explicitly opt-in, 3s-debounced code watcher with path-scoped reindex and shared-generation suppression
Write-guard snapshotting with rolling .bak backups (default: 5)
Isolated telemetry logging (~/.rms-memory/rms.log)
llms.txt export for flat, decoupled LLM ingestion

🧩 Supported IDEs

IDE	Auto-Install	Rules Injection
Cursor	✅	`.cursorrules`
Zed	✅	`.zed/assistant.md`
Claude Code	✅	`.claude/CLAUDE.md`
OpenCode	✅	—
Codex	✅	—
VS Code	✅	—
Antigravity	✅	—

📄 License

MIT License — see LICENSE for details.

Add this server

Host a copy with our keys and logs.

Server facts

Tools

Questions

How do I install Rms Memory?

Does Rms Memory need an API key?

What transport does Rms Memory use?

When was Rms Memory last checked?

Documentation

🧠 RMS Memory MCP

The Problem

🖥️ Prefer a GUI over raw Markdown?

✨ Key Features

📦 Installation

Option 1: Homebrew (macOS Apple Silicon & Linux)

Option 2: GitHub release binary

Option 3: Build from Source

crates.io (cargo install)

Optional RMS Memory GUI installers

🚀 Quick Start

Generated Wiki namespace

Use multiple isolated scopes

Configure your vault

Optional semantic code memory

Use an existing Obsidian vault

Embedding model and retrieval eval

🛠 CLI Commands

🔌 MCP Tools Exposed

🏗 Architecture Highlights

🧩 Supported IDEs

📄 License

Add this server

Host a copy with our keys and logs.

Server facts

Tools

Questions

How do I install Rms Memory?

Does Rms Memory need an API key?

What transport does Rms Memory use?

When was Rms Memory last checked?

Documentation

🧠 RMS Memory MCP

The Problem

🖥️ Prefer a GUI over raw Markdown?

✨ Key Features

📦 Installation

Option 1: Homebrew (macOS Apple Silicon & Linux)

Option 2: GitHub release binary

Option 3: Build from Source

crates.io (cargo install)

Optional RMS Memory GUI installers

🚀 Quick Start

Generated Wiki namespace

Use multiple isolated scopes

Configure your vault

Optional semantic code memory

Use an existing Obsidian vault

Embedding model and retrieval eval

🛠 CLI Commands

🔌 MCP Tools Exposed

🏗 Architecture Highlights

🧩 Supported IDEs

📄 License

crates.io (`cargo install`)

crates.io (`cargo install`)