Teepy MCP Server
Enables the Theopy AI assistant to interact with the Teepy pharmacy management ERP via the MCP protocol, with role-based access to customers, invoices, plannings, sessions, and reminders.
README
Teepy × Theopy — my contribution (sanitized extract)
This repository is a sanitized extract of Teepy, Kozea's proprietary pharmacy-management ERP (private repository, years of work by a full team, confidential to Kozea and its clients).
It exists to show, publicly and without breaching confidentiality, the code I personally wrote to connect Teepy to Theopy, an AI assistant I built as part of my RNCP39583 "Experte en Développement Logiciel" certification.
Every file and folder from the real Teepy repository is present here, at its real path, so the overall structure of the application is visible — but only the files I personally authored keep their real content. Everything else has been emptied (zero bytes) on purpose. This repo carries no git history from the original: only Teepy's own private repository has that.
Where my contribution sits in the architecture
┌─────────────────────────────────────────────────────-─────────────────┐
│ THEOPY (public, separate repo) │
│ AI assistant — Gemini / Ollama brain, chat UI, login │
└────────────────────────────────┬──────────────────────────────────────┘
│ MCP protocol over SSE
│ initialize → tools/list → tools/call
▼
┌──────────────────────────────────────────────────────────────────────┐
│ TEEPY (this repo — sanitized) │
│ │
│ ┌──────────────────────────────────────────────────────────────┐ │
│ │ mcp_server.py ★ MINE │ │
│ │ FastMCP entrypoint - boots the MCP server on :5001 │ │
│ └────────────────────────────┬─────────────────────────────────┘ │
│ │ │
│ ┌────────────────────────────▼───────────────────────────────--─┐ │
│ │ mcp_tools/ ★ MINE │ │
│ │ core.py - identity resolution via MCP `meta`, │ │
│ │ requires_role() access control │ │
│ │ invoices.py - factures (administrator/manager/commercial) │ │
│ │ customers.py - clients (administrator/manager/commercial) │ │
│ │ plannings.py - plannings (administrator/manager/operator) │ │
│ │ sessions.py - sessions (+ contractor) │ │
│ │ reminders.py - relances (administrator/manager/commercial) │ │
│ └────────────────────────────┬───────────────────────────────-──┘ │
│ │ calls agent_* functions │
│ ┌────────────────────────────▼──────────────────────────────-───┐ │
│ │ teepy/routes/theopy/ ★ MINE │ │
│ │ base.py, customer.py, invoice.py, planning.py, │ │
│ │ reminder.py, session.py │ │
│ │ → agent-facing business logic, wrapping Teepy's existing │ │
│ │ repositories/models │ │
│ └────────────────────────────┬─────────────────────────────-────┘ │
│ │ │
│ ┌────────────────────────────▼─────────────────────────-────────┐ │
│ │ teepy/routes/theopy_auth.py ★ MINE │ │
│ │ POST /api/theopy/authenticate │ │
│ └────────────────────────────┬───────────────────────────────-──┘ │
│ │ queries through │
│ ┌────────────────────────────▼─────────────────────────────-────┐ │
│ │ teepy/models/, teepy/repositories/, teepy/services/, │ │
│ │ teepy/routes/ (rest), teepy/templates/, teepy/static/, ... │ │
│ │ ○ PRE-EXISTING TEEPY │ │
│ │ proprietary ERP - content redacted in this repo │ │
│ └────────────────────────────┬───────────────────────────────-──┘ │
│ │ │
│ ┌────────────────────────────▼────────────────────────────-─────┐ │
│ │ PostgreSQL (Teepy production database) │ │
│ └────────────────────────────────────────────────────────────--─┘ │
└────────────────────────────────────────────────────────────────--────┘
★ MINE = full, real content, kept as-is.
○ PRE-EXISTING TEEPY = real file/folder structure preserved, content emptied.
What's mine, in full
| Path | What it is |
|---|---|
mcp_server.py |
FastMCP entrypoint for the MCP server |
mcp_tools/ (6 files) |
Role-based access control layer over every MCP tool |
teepy/routes/theopy/ (6 files) |
Agent-facing business logic (customers, invoices, plannings, reminders, sessions) |
teepy/routes/theopy_auth.py |
POST /api/theopy/authenticate - the endpoint Theopy's login calls |
test/theopy/ (7 files) |
Tests for the MCP tools and agent-facing functions, run against real business logic and a real database |
test/test_theopy_auth.py |
Tests for the authentication endpoint |
Makefile |
Includes commands I added: theopy-serve, mcp-inspect, install-db-test |
docker-compose.yml |
MCP server port (5001) and networking added for Theopy to reach Teepy |
sql/dev_data.sql |
Additional seed rows I added (synthetic dev data only, not real customer data) for realistic testing |
requirements/, uv.lock |
Dependency additions for the MCP integration |
teepy/__init__.py is intentionally emptied too, even though I added 2 lines
to it (registering the theopy_auth blueprint) - the file is overwhelmingly
pre-existing Teepy bootstrap code, not mine, so it follows the same rule as
everything else outside the list above.
Links
- Theopy (the AI assistant, fully public): https://github.com/SajedehAdelia/Theopy
- Teepy (the real repository, private - access on request): https://github.com/Kozea/teepy
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.