Marsnme

FreeNot checked

Agent-agnostic persistent memory backend. 13 MCP tools, Supabase + Jina embeddings, multi-profile isolation, semantic recall across sessions.

by Marsmanleo

GitHub Embed

About

Agent-agnostic persistent memory backend. 13 MCP tools, Supabase + Jina embeddings, multi-profile isolation, semantic recall across sessions.

README

English | 繁體中文（台灣） | 繁體中文（香港） | 简体中文

marsnme.com — Claude.md is for context. MarsNMe is for continuity.

Your AI tools should know you — not start from scratch every time. When Perplexity helps you decide, Claude should remember why. When Cursor ships a feature, Warp should know the context. That's not context sharing. That's continuity.

Most AI memory tools help AI remember you. MarsNMe helps you and your AI remember each other — across sessions, across tools, over time.

An agent-agnostic, LLM-agnostic memory backend for MCP-compatible tools.

curl -fsSL https://marsnme.com/install.sh | bash

MarsNMe on Glama npm version MCP Registry LobeHub npm downloads License GitHub stars

Real user story

I use Cursor to code, Warp to deploy, Perplexity to research, and Claude Code to manage my vault. Before MarsNMe, every tool started blank — I had to re-explain my project, my preferences, my decisions every single session. Now my AI across all four tools knows what we decided yesterday, what we tried last week, and why we chose this architecture over that one. It's not about injecting context. It's about having a relationship that compounds over time.

— Leo, MarsNMe creator (3 months of daily use across 4 AI tools)

Available MCP Tools (16)

Tool	Description
`insert_memory`	Store short-term memory
`list_memories`	List recent memories
`search_memories`	Semantic search via Jina embeddings
`recall`	Long-term chunk recall — ~80-char preview per match
`get_summary`	Medium excerpt (~300 chars) of a chunk by ID
`get_full`	Complete text of a long-term chunk by ID
`memory_ingest`	Ingest long-term insight chunks
`dream_ingest`	Dream-mode long-term ingestion
`session_boot`	Start a session with context pre-load
`session_close`	Close session, summarize, auto-promote expiring memories
`health_check`	Coverage, expiry, conflict diagnostics
`reload_source_registry`	Refresh source whitelist at runtime
`demote_memory`	Demote a memory to lower priority
`soft_forget`	Soft-delete a memory
`explain_memory`	Explain a memory's provenance
`batch_promote`	Promote expiring short-term memories to long-term

What's new in 0.3.0

3-layer recall: recall returns ~80-char previews, then get_summary (~300 chars), then get_full (complete). Avoids token-dumping full chunks on every recall; drill down only when a preview looks relevant.
Body-to-body note handoff: session_close(to=<body>, note=...) leaves a note that session_boot(body=<target>) delivers and marks read — one agent can hand context to another.
Auto batch_promote on session_close: closing a session automatically promotes soon-expiring short-term memories (48h window, up to 5) to long-term storage — no Hermes or manual promote needed.
grok + draft sources: added to the source whitelist so the Grok body and Draft lifecycle hooks can write memories natively.
CoCo-only tool surface: PRD tools (save_prd, get_prd, list_prds, score_prd, spawn_to_linear) removed from the Supabase gateway — idea/PRD/task execution now lives in Draft. MarsNMe = CoCo soul memory only.

MarsNMe

Why MarsNMe?

Most AI memory tools help AI remember you. MarsNMe helps you and your AI remember each other.

	MarsNMe	Typical memory tool
Philosophy	Mutual continuity — human + AI both grow	AI-side context injection only
Agent support	Any MCP-compatible client	Often client-specific
Memory tiers	Short-term (TTL) + long-term (semantic)	Usually one layer
Profiles	Unlimited isolated profiles via `MCP_PROFILE`	Single-user only
Data ownership	Your own Supabase — zero vendor lock-in	Vendor-hosted
Search	Jina v3 semantic search (1024-dim pgvector)	Keyword or basic similarity
Self-hostable	✅ Full control	Rarely

When MarsNMe is the right fit

You use multiple AI assistants (Claude, Cursor, Perplexity, Warp, custom agents) and want shared memory across all of them
You want AI that remembers your projects, preferences, and decisions across sessions without re-explaining
You care about data sovereignty — your memories stay in your own Supabase project
You're building an AI agent and need a production-ready memory backend with semantic recall

When it might not be the right fit

You only need single-session context (just use the system prompt)
You want fully managed, zero-config memory with no setup (try a hosted solution)

Runtime packages

Folder	Runtime	Who uses it
`marsnme-supabase/`	Supabase + Jina gateway (`@marsnme/mcp-gateway`)	Mars Group dogfood — Proxmox CT101 (CoCo / Toto soul memory)
`marsnme-cf/`	Cloudflare Workers + D1 + Vectorize	Self-host template; not the Proxmox deploy path
`marsnme-supabase/cloudflare-routing-worker/`	`mcp.marsnme.com` routing proxy	Public setup wizard → upstream gateway

Product split (Mars Group): Idea / PRD / task execution → Draft + draft-mcp. MarsNMe Supabase = CoCo soul memory only (recall, session boot/close, ingest, lifecycle). As of @marsnme/mcp-gateway v0.3.0, PRD MCP tools (save_prd, get_prd, list_prds, score_prd, spawn_to_linear) are removed from the Supabase gateway — use Draft for idea/PRD/task workflows.

Proxmox deploy: private MarsNMe-lab — deploy/deploy-proxmox-ct101.sh or GitHub cd-selfhosted workflow. Not a single-script deploy like draft-mcp.

Repository Packages

Package	Description
`marsnme-supabase/`	Core MCP gateway — agent-agnostic memory backend (this package is published to npm as `@marsnme/mcp-gateway`)
`marsnme-supabase/cloudflare-routing-worker/`	Cloudflare Worker for `mcp.marsnme.com` — username-based MCP routing proxy with setup wizard
`marsnme-cf/`	Self-hosted MCP memory server on Cloudflare Workers + D1 + Vectorize (no Supabase needed)

Quick Setup (no install needed)

Go to mcp.marsnme.com/setup — create your personal MCP URL in 4 steps:

Pick a username
Enter your Supabase credentials (URL + anon key)
Choose preferences
Get your MCP URL: https://mcp.marsnme.com/your-name

Then add it to any MCP client (Claude, Cursor, Perplexity, Warp).

Self-hosted? Deploy marsnme-cf/ to your own Cloudflare account — no Supabase needed, uses D1 + Workers AI + Vectorize.

Before You Start (External Dependencies)

Create a Supabase project (free plan is enough):
- Sign up: https://supabase.com
- Create project: https://supabase.com/dashboard/new
- Open API settings (Project Settings → API):
  - Project URL → SUPABASE_BASE_URL
  - service_role key → SUPABASE_SERVICE_ROLE_KEY
- Keep SUPABASE_SERVICE_ROLE_KEY private. Never commit it.
Create a Jina API key (free tier available):
- Get key: https://jina.ai/api-key/
- Copy key to JINA_API_KEY

Quick Start (15-20 minutes)

For the fastest path, use the one-line installer: curl -fsSL https://marsnme.com/install.sh | bash

The manual path below follows the same tools-first flow as docs/onboarding-a-mcp-zero-to-recall.md and docs/onboarding-b-platform-skill-install.md.

Clone repository:

git clone https://github.com/Marsmanleo/MarsNMe.git
cd MarsNMe

Verify Node.js version (20+ required):

node --version

Copy environment template:

cp .env.example .env

Fill required values in .env:
- SUPABASE_BASE_URL
- SUPABASE_SERVICE_ROLE_KEY
- JINA_API_KEY
Run required Supabase migrations before first start:
- Option A (recommended, Supabase CLI):

npx supabase db push --db-url "<your-supabase-db-connection-string>"

Note: --db-url must be the Postgres database connection string from Project Settings → Database → Connection string.
It is not the same as SUPABASE_BASE_URL (https://<project-ref>.supabase.co, REST API URL).
Use a role that can execute DDL on your target schemas.
On Supabase-hosted Postgres this is typically supabase_admin (not postgres).
Option B (Supabase Dashboard SQL Editor):
1. Open SQL Editor.
2. Ensure the vector extension is enabled first (Database → Extensions).
3. Run migration files in filename order from supabase/migrations/:
  - 20260504052744_semantic_vector_dual_profile.sql
  - 20260513213800_memory_lifecycle_tracking.sql
  - 20260513222500_health_check_detect_conflicts_v2.sql
  - 20260517183000_provenance_audit_trail.sql
  - 20260517194000_memory_scope_agent_body_environment.sql
  - 20260517200500_forget_demote_mechanism.sql
  - 20260517223500_usage_cost_telemetry_light.sql
  - 20260517231000_memories_source_constraint_regex.sql
  - 20260517232000_source_registry_table.sql

Start gateway:
- MCP_PROFILE separates memory by agent or use case.
- Use any profile name you want (for example: default, my-agent, profile-a).
- Legacy built-in profile IDs coco and toto are still supported for compatibility.
- If PORT is omitted, default port is profile-based (coco=18790, toto=18791, other profiles deterministic in 20000-29999).

MCP_PROFILE=profile-a PORT=18790 npx @marsnme/mcp-gateway

Verify health:

curl -sS http://127.0.0.1:18790/health

Connect your MCP client (next section), then run the first round-trip check.

Try In 30 Seconds (Docker, M1)

If you only want a local demo path, use Docker Compose.

One-line install (recommended):

curl -fsSL https://marsnme.com/install.sh | bash

Or manually:

Set only the required key:

cp .env.example .env
# fill JINA_API_KEY in .env

Start local stack:

docker compose up

This starts:

PostgreSQL + pgvector
SQL migrations from supabase/migrations/
PostgREST + rest-proxy
MarsNMe gateway (http://127.0.0.1:18790/mcp)

Verify health:

curl -sS http://127.0.0.1:18790/health

M2 Cloudflare Tunnel Profile (Demo)

When you need a temporary public endpoint for remote AI tools:

docker compose --profile tunnel up

Expected output (from tunnel logs):

https://xxxx.trycloudflare.com

Get MCP endpoint:

docker compose --profile tunnel logs tunnel | grep -Eo 'https://[^ ]+trycloudflare.com' | head -n1
# append /mcp

Notes:

trycloudflare.com URL is temporary (demo only).
Local endpoint remains: http://127.0.0.1:18790/mcp.
For production/stable URL, use named tunnel (outside M2 scope).
Optional env:
- MCP_TUNNEL_PROFILE (default coco)
- MCP_TUNNEL_REQUIRE_BEARER (default false for demo convenience)

MCP Client Connection Guide

Local endpoint:

http://127.0.0.1:18790/mcp

If bearer auth is enabled (MCP_REQUIRE_BEARER=true), include:

Authorization: Bearer <your-token>

Claude Desktop

Open claude_desktop_config.json (macOS default path: ~/Library/Application Support/Claude/claude_desktop_config.json).
Add/update:

{
  "mcpServers": {
    "marsnme-cf": {
      "url": "http://127.0.0.1:18790/mcp"
    }
  }
}

Restart Claude Desktop.

Cursor

Open Cursor Settings and search for MCP.
Add a new server:
- Name: marsnme-cf
- URL: http://127.0.0.1:18790/mcp
- Headers: optional bearer header if enabled
Reconnect MCP in Cursor.

Warp

Open Settings > Agents > MCP servers.
Add a server pointing to:
- URL: http://127.0.0.1:18790/mcp
Add optional bearer header if required, then reconnect.

Perplexity

Open a Space in Perplexity and go to Space Settings.
Under MCP servers, add:
- URL: http://127.0.0.1:18790/mcp
Save and start a new conversation in that Space.

Any MCP client (generic HTTP/SSE)

Use a streamable HTTP/SSE MCP entry:

{
  "marsnme-cf": {
    "url": "http://127.0.0.1:18790/mcp"
  }
}

First Connection Validation (Round Trip)

After client connection, verify this sequence once:

tools/list:

curl -sS http://127.0.0.1:18790/mcp \
  -H 'content-type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'

insert_memory:

curl -sS http://127.0.0.1:18790/mcp \
  -H 'content-type: application/json' \
  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"insert_memory","arguments":{"body":"quickstart memory check","source":"warp","session_id":"quickstart-smoke"}}}'

recall:

curl -sS http://127.0.0.1:18790/mcp \
  -H 'content-type: application/json' \
  -d '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"recall","arguments":{"query":"quickstart memory check","limit":3}}}'

What this repository is

mars-memory-mcp is the core MCP gateway repository behind the public-facing MarsNMe release.
One codebase (marsnme-supabase/server.mjs) serves multiple profile schemas through MCP_PROFILE. This public repository currently keeps two built-in legacy profile IDs (coco, toto) for backward compatibility.

Current capabilities

MCP methods: initialize, notifications/initialized, tools/list, tools/call, ping
Profiles: configurable profile IDs (legacy built-ins: coco, toto)
Memory tools (16):
- insert_memory (short-term memory)
- list_memories
- search_memories (Jina embedding search)
- recall (~80-char preview) then get_summary (~300-char excerpt) then get_full (complete text)
- memory_ingest / dream_ingest (long-term chunk ingestion)
- session_boot / session_close (daily rhythm lifecycle; close auto-promotes expiring memories + supports body-to-body note handoff)
- health_check (coverage, expiry, conflict diagnostics)
- reload_source_registry (refresh source whitelist at runtime)
- demote_memory / soft_forget / explain_memory (memory lifecycle management)
- batch_promote (promote expiring short-term memories to long-term)
Sources: perplexity, cursor, warp, openclaw, hermes, draft, grok
OAuth-protected MCP endpoint (configurable by environment variables)

Memory model

Short-term memory table: <profile>.memories
Long-term memory table: <profile>.marsvault_chunks
Recommended usage:
- Keep daily interaction context in insert_memory
- Promote durable insights through ingest tools

Repository layout

marsnme-supabase/server.mjs — gateway entry point
marsnme-supabase/scripts/hermes_digest_runner.py — optional digest runner
marsnme-supabase/scripts/dream_runner.py — public self-host dream runner
marsnme-supabase/deploy/systemd/ — systemd templates
marsnme-supabase/deploy/phase2/ — build/deploy scripts
marsnme-supabase/deploy/phase3/smoke_gate.sh — smoke gate script
supabase/migrations/ — schema-as-code migrations

Environment setup

Copy .env.example to your local .env (do not commit real secrets).
Fill required values:
- MCP_PROFILE (your profile identifier; this repo ships with legacy coco/toto)
- SUPABASE_BASE_URL
- SUPABASE_SERVICE_ROLE_KEY
- JINA_API_KEY
Optional security flags:
- MCP_REQUIRE_BEARER=true
- MCP_CLIENT_ID
- MCP_CLIENT_SECRET

Optional Hermes digest runner

Hermes is optional and disabled by default:

HERMES_ENABLED=false
HERMES_DIGEST_MCP_URL
HERMES_DIGEST_MCP_BEARER_TOKEN
HERMES_DIGEST_ORIGIN
HERMES_DIGEST_SOURCE_DIR

Optional Dream Runner (self-host)

Dream Runner is public-friendly and can run without Hermes private environment:

DREAM_ENABLED=true
DREAM_MODE=lite|standard|pro
DREAM_DIGEST_MCP_URL
DREAM_MCP_BEARER_TOKEN (if required)
DREAM_ENABLE_ISSUE_SIGNALS, DREAM_ENABLE_REPO_SCAN, DREAM_ENABLE_SOUL_CONTEXT (optional overrides)

Quick start:

DREAM_ENABLED=true DREAM_MODE=lite python3 marsnme-supabase/scripts/dream_runner.py

If you run this repository with bundled defaults and no profile remapping, use coco and toto.

See docs/dream-runner-self-host.md for full configuration.

Onboarding

Zero-to-first-recall guide: docs/onboarding-a-mcp-zero-to-recall.md
Platform install guide (optional skill layer): docs/onboarding-b-platform-skill-install.md

Skill library

Skill index and update workflow: skills/README.md
Perplexity template: skills/perplexity/memory-daily-boot/SKILL.md
Cursor template: skills/cursor/memory-daily-boot/rule.mdc
Warp template: skills/warp/memory-daily-boot/prompt.md

Local run (from cloned repo)

MCP_PROFILE=profile-a npx @marsnme/mcp-gateway

MCP_PROFILE=profile-b npx @marsnme/mcp-gateway

Health endpoints:

GET /health
POST /mcp

Systemd deployment

Use marsnme-supabase/deploy/systemd/[email protected] with instances:

[email protected]
[email protected]

Recommended env files:

/opt/mars-memory-mcp/shared/.env
/opt/mars-memory-mcp/shared/.env.profile-a
/opt/mars-memory-mcp/shared/.env.profile-b

Release/deploy scripts

Build artifact:

bash marsnme-supabase/deploy/phase2/build_release_artifact.sh

Apply migrations with an explicit DDL-capable role:

npx supabase db push --db-url "<postgres://supabase_admin:<password>@<host>:5432/postgres>"

Run pre-deploy schema gate (must pass before any service restart):

bash marsnme-supabase/deploy/phase2/pre_deploy_schema_gate.sh \
  --db-url "<postgres://supabase_admin:<password>@<host>:5432/postgres>" \
  --profiles coco,toto \
  --expected-role supabase_admin

Run your platform-specific rollout/restart adapter.
- This repository ships generic artifact + gate scripts; rollout adapters are environment-specific.
- If schema gate exits non-zero, stop deployment and do not restart services.
Smoke gate:

bash marsnme-supabase/deploy/phase3/smoke_gate.sh --spawn-local

Automated npm + MCP Registry release (tag-driven):
- Workflow: .github/workflows/publish-release.yml
- Trigger: push tag v*
- Gate: tag version must match marsnme-supabase/package.json version
- Optional local Fish helper:

mrel patch
mrel minor
mrel major
mrel 0.1.2

The helper updates marsnme-supabase/package.json and server.json, commits, tags, and pushes.

Security and version control

Never commit .env, runtime tokens, or oauth-clients.json
Keep .env.example committed as the only environment template
Prefer bearer/OAuth for public exposure

License and policy

License: Apache-2.0 (LICENSE)
Notice: NOTICE
Trademark policy: TRADEMARK.md
Contribution guide: CONTRIBUTING.md
Contributor agreement: CLA.md
Release notes: CHANGELOG.md

from github.com/Marsmanleo/MarsNMe

Install Marsnme in Claude Desktop, Claude Code & Cursor

Recommended · one command, every IDE

unyly install marsnme

Installs into Claude Desktop, Claude Code, Cursor & VS Code — handles npx, uvx and build-from-source repos for you.

First time? Get the CLI: curl -fsSL https://unyly.org/install | sh

Or configure manually

Run in your terminal:

claude mcp add marsnme -- npx -y @marsnme/mcp-gateway

FAQ

Is Marsnme MCP free?

Yes, Marsnme MCP is free — one-click install via Unyly at no cost.

Does Marsnme need an API key?

No, Marsnme runs without API keys or environment variables.

Is Marsnme hosted or self-hosted?

A hosted option is available: Unyly runs the server in the cloud, no local setup required.

How do I install Marsnme in Claude Desktop, Claude Code or Cursor?

Open Marsnme on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install — the config is generated automatically, no JSON editing.

Related MCPs

Fetch

Web content fetching and conversion for efficient LLM usage.

by Community

AWS KB Retrieval

Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.

by modelcontextprotocol

Spring AI MCP Server

Provides auto-configuration for setting up an MCP server in Spring Boot applications.