coach-memory
A personal career-coach memory service that ingests coaching logs, embeds them for semantic search, and exposes the memory to Claude clients via MCP, with a Next.js dashboard for browsing.
README
Personal Coach
A personal career-coach memory service. It ingests daily/weekly/quarterly coaching logs and notes, embeds them into Postgres (pgvector) for hybrid semantic + keyword search, and exposes that memory to Claude clients over an MCP server behind BetterAuth (OAuth + API keys). It also serves a small Next.js dashboard (timeline, search, notes capture, analytics) for browsing the same data directly.
Prerequisites
- Node.js 22+
- Docker (for the local Postgres/pgvector instance)
Local setup
- Start Postgres:
docker compose up -d - Copy/create
.env.localwith the required env vars (see below). - Apply the schema (coach tables + better-auth tables):
npm run migrate - Create a login user and a sync API key:
ALLOW_SIGNUP=true npm run create-user -- <email> <password> npm run create-api-key -- sync-cli
Environment variables
| Variable | Used by | Purpose |
|---|---|---|
DATABASE_URL |
app, migrate, scripts | Postgres connection string |
OPENAI_API_KEY |
app | Embeddings for ingest/search |
BETTER_AUTH_SECRET |
app | BetterAuth session/token signing secret |
BETTER_AUTH_URL |
app | Public base URL BetterAuth issues callbacks against |
COACH_MEMORY_URL |
sync CLI | Base URL of the Personal Coach server to sync against |
COACH_SYNC_API_KEY |
sync CLI | API key (from create-api-key) used to authenticate sync pushes/pulls |
Scripts
npm test # run the test suite (vitest, against .env.test)
npm run build # production build (next build)
npm run sync # push local docs, pull notes mirror, watermark progress
Production deploy
Coolify (recommended — one resource, no manual proxy/DB wiring)
docker-compose.coolify.yml deploys the app and its pgvector database together as a
single "Docker Compose" resource:
- In Coolify: New Resource → Docker Compose, point it at this repo, and set the
Compose file path to
docker-compose.coolify.yml. - Set two environment variables in the Coolify UI:
OPENAI_API_KEYandBETTER_AUTH_SECRET(openssl rand -hex 32). Everything else — the DB credentials, the app's public domain and TLS cert — is generated and wired automatically via Coolify'sSERVICE_FQDN/SERVICE_URLmagic variables. - Deploy. The schema applies itself on every container boot (see
scripts/docker-entrypoint.mjs), so there's no separate migrate step. - One-time bootstrap: temporarily expose the
dbservice's port (or use an SSH tunnel to the Coolify host), then from your local checkout run against thatDATABASE_URL:ALLOW_SIGNUP=true npm run create-user -- <email> <password> npm run create-api-key -- sync-cli - Connect a Claude client to
https://<the-domain-coolify-assigned>/mcp.
Any other container host
The plain Dockerfile at the repo root also builds a standalone image on its own —
point it at any pgvector/pgvector Postgres instance, set the environment variables
above, and front it with your own TLS/reverse-proxy setup (e.g. a Cloudflare
Tunnel). The image still self-migrates on boot, so no separate npm run migrate
step is needed; only the one-time user/API-key bootstrap above applies.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.