From nothing to agents that remember

Part 1

Memory, in one command

1. Step 1

   Install and run setup

   Setup asks a few questions (database, business name, your name, template, which tools to connect), shows what it is about to change, and waits for a yes before each part. It finishes by writing one fact and recalling it.

   npm install -g @mycobrain/mcp-server
   mycobrain-setup memory

   Database: setup uses a local Postgres with pgvector if it finds one, or offers to start one in Docker. If neither is there, it prints the install commands and changes nothing.

   Templates: agency, design studio, architecture or landscape firm, solo consultant, or custom (a short guided interview).

   Claude Code and Codex get hooks: recall at session start and on each prompt, and capture of finished sessions. Cursor, Windsurf and Claude Desktop get an MCP server with the two memory tools.

   Each changed file is backed up first, nothing secret is written into it, and running setup again adds nothing twice.

2. Step 2

   Or use Supabase, Neon, RDS, or any Postgres

   Put the connection string in an environment variable and pass a reference to it, so it never lands in your shell history. On Supabase, use the direct connection or the session pooler on port 5432.

   export MYCOBRAIN_DATABASE_URL='postgresql://...'
   mycobrain-setup memory --database url --database-url-ref env:MYCOBRAIN_DATABASE_URL

3. Step 3

   Add your clients

   Houses keep clients apart. Entities are the people, projects and companies inside them. Only you create them, so an agent can never widen its own scope.

   mycobrain-memory house add riverton-parks "City of Riverton Parks"
   mycobrain-memory entity add "Riverside Park renovation" --kind project --house riverton-parks

   When a fact changes, the new one supersedes the old one. mycobrain-memory history <ref or id> shows both and the reason.

4. Step 4

   Check it, any time

   The doctor reports the config, the database, each tool's credential, the hooks, the background worker and Ollama, with the fix for anything that is off.

   mycobrain-doctor --memory

Part 2, optional

Add the document engine

The engine ingests files, folders and repos, builds a knowledge graph, and grows its own schema. One docker compose up boots it, with no API keys.

1. Step 1

   Boot the stack

   Clone the repo and bring everything up with Docker Compose. Three services start: Postgres 16 + pgvector, the MCP server, and the extraction worker, with a seeded default workspace, so it works with zero configuration.

   git clone https://github.com/thegoodguysla/myco-brain.git
   cd myco-brain
   docker compose up -d

   Verify with docker compose ps.

2. Step 2

   See it work in 30 seconds

   Ingest the bundled demo corpus, a small set of interconnected documents for a fictional agency and its client. Full-text (BM25) search and ingestion are keyless, so no API key is needed.

   npx -y -p @mycobrain/mcp-server mycobrain-ingest ./examples/demo-corpus

   After connecting your client in the next step, ask: “When does Northwind’s rebrand launch, and who owns the account?”

   Follow up with: “Show me the source for that.”

3. Step 3

   Connect Claude Desktop

   Add Myco Brain to ~/Library/Application Support/Claude/claude_desktop_config.json, then restart Claude Desktop. Using Cursor, Windsurf, Continue, or Zed instead? Point your client’s MCP settings at the same command and env vars.

   {
     "mcpServers": {
       "myco-brain": {
         "command": "npx",
         "args": ["-y", "@mycobrain/mcp-server"],
         "env": {
           "DATABASE_URL": "postgresql://brain:brain@localhost:5432/brain",
           "BRAIN_WORKSPACE_ID": "00000000-0000-0000-0000-000000000001",
           "BRAIN_API_KEY": "brain_00000000-0000-0000-0000-000000000001_00000000-0000-0000-0000-0000000000a1_localdev"
         }
       }
     }
   }

   These exact values work for everyone, on purpose. localhost is your own machine (the Docker stack from the first engine step), and the workspace ID and API key are local-dev defaults seeded into every fresh install. Your database, on your computer; nothing is shared between installs.

   Happy-path test: save a memory in one session (“remember that the board meeting is every Wednesday at 9 AM Pacific”), open a fresh session, and ask for it back.

4. Step 4

   Bulk ingest your own knowledge

   Point the ingest CLI at a local folder or a whole GitHub repo. It skips binaries and noise directories like node_modules and .git automatically.

   export DATABASE_URL=postgresql://brain:brain@localhost:5432/brain
   export BRAIN_WORKSPACE_ID=00000000-0000-0000-0000-000000000001
   export BRAIN_API_KEY=brain_00000000-0000-0000-0000-000000000001_00000000-0000-0000-0000-0000000000a1_localdev
   
   npx -y -p @mycobrain/mcp-server mycobrain-ingest ./docs
   npx -y -p @mycobrain/mcp-server mycobrain-ingest github:owner/repo

   Set GITHUB_TOKEN to ingest private repos.