Experiments over agentic approaches with LLM. This is an educative project which selectively tries to avoid more structured frameworks.
It uses MCP as repositories for tools, prompts and resources.
In ./config there is a sample YAML to configure agents and tasks to perform to develop a TODO list web application and a SOTA for REC in Europe.
Note: This project uses Task for automation, consider installing to ease the setup.
- Install dependencies with a package manger e.g.
npm i - Build the base image for the MCP proxy
docker build ./mcp/mcp-proxy-base -f ./mcp/mcp-proxy-base/Dockerfile -t mcp-proxy/base
docker build ./mcp/mcp-servers -f ./mcp/mcp-servers/Dockerfile.base -t mcp-proxy/servers-base- Create a
.envin the root folder and set the those api keys
# openai api key
OPENAI_API_KEY=...
# serper.dev api key
SERPER_API_KEY=...
- Add or edit tasks definition in
./config
Start the MCP servers to be used by the agents.
docker compose -f docker-compose.mcp.yaml up -d
Start the main runtime with docker compose up
Copyright 2025 Luca Capra
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.