Skip to main content
Glama
tswic-highdesert

Firmament Open Brain

Firmament Open Brain

A shared "context brain" for the Firmament team, inspired by OB1 but rebuilt lean for Railway. It's an MCP (Model Context Protocol) server backed by Postgres: each person gets a private memory store that follows them across AI tools (Claude Desktop, Claude Code, claude.ai, Cursor, etc.). Memories are isolated per user — your token only ever sees your own memories.

How it works

  • One Railway service (this Node app) + one Railway Postgres.

  • Each user gets a personal token. Their MCP URL is https://<host>/mcp/<token>.

  • All memory reads/writes are scoped to the user that token resolves to. Tokens are stored hashed (SHA-256).

  • Search is Postgres full-text search with a substring fallback — no embedding API keys needed.

Related MCP server: LedgerMem MCP Server

MCP tools exposed

Tool

What it does

remember

Store a memory (content + optional tags)

recall

Search your memories by topic/keywords

recent_memories

List newest memories

forget

Delete a memory by id

Connecting a client

Claude Code:

claude mcp add --transport http openbrain https://<host>/mcp/<your-token>

Claude Desktop / claude.ai: Settings → Connectors → Add custom connector → paste https://<host>/mcp/<your-token>.

Cursor: Settings → MCP → Add server, type http, same URL.

Then just talk: "remember that ..." / "what do you know about ...". A good habit is to start sessions with "check my open brain for context on ".

Admin (minting users)

Requires the ADMIN_TOKEN env var; pass it as a Bearer token.

# create a user (returns their token + MCP URL — shown only once)
curl -X POST https://<host>/admin/users \
  -H "Authorization: Bearer $ADMIN_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "Jack Wright"}'

# list users
curl https://<host>/admin/users -H "Authorization: Bearer $ADMIN_TOKEN"

# delete a user (and all their memories)
curl -X DELETE https://<host>/admin/users/<id> -H "Authorization: Bearer $ADMIN_TOKEN"

Deploy (Railway)

Env vars on the app service:

  • DATABASE_URL — reference the Railway Postgres

  • ADMIN_TOKEN — long random string (admin API auth)

  • DATABASE_SSL — set to false if connecting over Railway private networking

Deploy with railway up from this directory. Schema auto-creates on boot.

Security notes

  • The token lives in the URL. That's what makes hookup one-paste-easy, and it's acceptable for a small trusted team, but treat the URL like a password: don't share it, don't paste it in group chats.

  • Lost token → delete the user, mint a new one (memories are keyed to the user row, so deleting the user deletes their memories — export first if needed).

F
license - not found
Not graded
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

View all related MCP servers

Related MCP Connectors

  • Shared, governed long-term memory for AI agents across tools and sessions via MCP and REST.

  • Your memory, everywhere AI goes. Build knowledge once, access it via MCP anywhere.

  • Person-owned AI memory that learns, not just stores — portable context for any MCP client.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/tswic-highdesert/openbrain-railway'

If you have feedback or need assistance with the MCP directory API, please join our Discord server