kirundi-brain-mcp
A FastMCP server that serves as a local-first Kirundi study coach, with tools to manage lesson files, track learner profiles, and provide grounded tutoring and answer grading via OpenAI.
README
Kirundi Study Brain (Week 6 MCP)
Local-first Kirundi (Rundi) study coach exposed as a FastMCP server. An OpenAI Agents SDK demo attaches over stdio, calls path-jailed Brain tools, and tutors from a Markdown corpus — not from thin air.
Inspired by the course trading/MCP labs and pro patterns like asket-mcp (Brain + profile + gated writes), scoped to Burundi language learning.
Architecture
demo_agent.py → MCPServerStdio → server.py (FastMCP)
│
┌────────────────────┼────────────────────┐
▼ ▼ ▼
brain/*.md SQLite profile ask / grade
(path jail) (data/) (OpenAI + retrieve)
Tools
| Tool | Purpose |
|---|---|
brain_list_files |
List corpus entries |
brain_read_file |
Read a lesson |
brain_search_notes |
Keyword search in .md |
brain_write_markdown |
Add/update lesson (overwrite gated) |
brain_delete_file |
Delete lesson (confirm gated) |
profile_get / profile_update |
Learner level, goals, mistakes |
ask_kirundi_tutor |
Retrieve + grounded answer |
grade_kirundi_answer |
Score a short reply |
kirundi_brain_info |
Paths / version / config |
Setup
cd 6_mcp/community_contributions/Sama-ndari_kirundi-brain
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env # set OPENAI_API_KEY
Run MCP server alone (stdio — Cursor / Claude Desktop)
python server.py
Example Cursor mcp.json fragment (adjust absolute paths):
{
"mcpServers": {
"kirundi-study-brain": {
"command": "python",
"args": ["/ABS/PATH/Sama-ndari_kirundi-brain/server.py"],
"env": {
"OPENAI_API_KEY": "sk-...",
"KIRUNDI_BRAIN_DIR": "/ABS/PATH/Sama-ndari_kirundi-brain/brain",
"KIRUNDI_BRAIN_DATA_DIR": "/ABS/PATH/Sama-ndari_kirundi-brain/data"
}
}
}
}
Run agent demo
python demo_agent.py "How do I say thank you in Kirundi?"
python demo_agent.py "Quiz me on greetings"
Sample corpus
Bundled under brain/: greetings, courtesy, numbers, food. Point KIRUNDI_BRAIN_DIR at a larger phrase bank when you have one.
Env
| Variable | Role |
|---|---|
OPENAI_API_KEY |
Required for agent demo + ask / grade |
KIRUNDI_BRAIN_MODEL |
Default gpt-4o-mini |
KIRUNDI_BRAIN_DIR |
Corpus root (default: ./brain) |
KIRUNDI_BRAIN_DATA_DIR |
SQLite dir (default: ./data) |
Do not commit .env or data/*.sqlite3.
Author
Sama-ndari — https://www.samandari.dev
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.