edge-ai-MCP
Enables ArduPilot drone control through MAVSDK plugins, supporting mission upload, telemetry, and flight commands via MCP tools.
README
edge-ai-MCP
MAVSDK-only drone backend for the SAR stack. All flight I/O goes through MAVSDK (Action, MissionRaw, Telemetry). No raw MAVLink.
Used internally by gateway → frontend LLM via a drone-http compatible HTTP API on port 3001.
SAR stack (recommended)
cd /home/jetson/Code
./sar-stack.sh start proxy # or: auto | serial
The drone tmux window runs scripts/stack_server.py:
| Port | Purpose |
|---|---|
| 3001 | Gateway LLM tools + frontend telemetry (DRONE_SERVER_URL) |
Flow:
Frontend chat → gateway :3000/infer → LLM → POST :3001/v1/apply-tool → MAVSDK → FC
Prerequisites
cd MCP
pip install -r requirements.txt
Environment
| Variable | Default | Purpose |
|---|---|---|
MAVSDK_CONNECT |
udpin://0.0.0.0:14550 |
MAVSDK connection |
MCP_HTTP_PORT |
3001 |
HTTP API for gateway |
MCP_MIN_ALT_M |
2 |
Min altitude (m) |
MCP_MAX_ALT_M |
120 |
Max altitude (m) |
MCP_MAX_DISTANCE_M |
2000 |
Max horizontal distance from home (m) |
MCP_MAX_SPEED_M_S |
15 |
Max speed for offboard velocity |
MCP_MAX_WAYPOINTS |
120 |
Max waypoints per mission upload |
Test from frontend chat
With sar-stack running, try in the dashboard:
- "Just hover in place for now"
- "Return to home immediately" (SITL / when safe)
Test scripts
python3 scripts/test_unit.py
python3 scripts/test_sitl.py --skip-flight
Mission upload (dashboard Mission Planner)
{
"include_takeoff": true,
"takeoff_alt_m": 15,
"include_rtl": true,
"waypoints": [
{"lat_deg": 23.558, "lon_deg": 120.473, "alt_m": 15}
]
}
Posted to POST /v1/mission/upload (proxied by gateway as /drone/mission/upload).
Standalone stdio MCP (optional, local dev only)
Not started by sar-stack. For Cursor/debug on the Jetson:
python3 server.py
Deploy
git remote -v # git@github.com-edge-ai-mcp:basilrari/edge-ai-MCP.git
git push -u origin main
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.