devto-agent
Build and deploy an autonomous Devto Agent capable of interacting with the Dev.to platform, powered by A2A (Agent-to-Agent) and MCP (Model Context Protocol)
claude mcp add --transport stdio heetvekariya-devto-agent uv run mcp_servers/sse/devto_server.py \ --env DEVTO_API_KEY="your_devto_api_key_here" \ --env DEVTO_BASE_URL="https://dev.to/api" \ --env GOOGLE_API_KEY="your_google_api_key_here"
How to use
DevTo Agent provides a multi-agent workflow that connects A2A agents with MCP servers to manage Dev.to content and user data. The MCP layer exposes a Server-Sent Events (SSE) interface for real-time, event-driven communication and a STDIO interface for straightforward command-driven interactions. Through the DevTo service wrapper, the system can fetch articles, post markdown content, retrieve user profiles, and manage reading lists. Tools connect the A2A agents to these MCP servers, enabling automated article creation, tagging, and user actions via programmable commands. You can use the SSE endpoint to subscribe to live updates and issue commands via the /messages/ endpoint, while the STDIO setup offers a simpler, script-friendly mode for batch operations and testing.
How to install
Prerequisites:
- Python 3.10 or higher
- UVicorn-compatible runtime (uv) for running MCP/Agent servers
- DevTo API key and Google API key
Installation steps:
-
Clone the repository and install dependencies: git clone https://github.com/HeetVekariya/devto-agent.git cd devto-agent
-
Create environment configuration file with required keys: DEVTO_API_KEY=your_devto_api_key_here DEVTO_BASE_URL=https://dev.to/api GOOGLE_API_KEY=your_google_api_key_here
-
Install Python package in editable mode (from project root): uv install -e . (or: python -m pip install -e .)
-
Start the MCP SSE server (DevTo MCP server): uv run mcp_servers/sse/devto_server.py
-
Start the A2A DevTo agent server: uv run a2a_servers/agent_server/devto_agent.py
-
Start the Host Agent (if using multi-agent orchestration): uv run a2a_servers/agent_server/host_agent.py
-
Run the client application (optional): uv run main.py
Notes:
- Ensure the environment variables are loaded in the environment where UV run executes.
- If you use a virtual environment, activate it prior to installation and startup.
Additional notes
Tips and common issues:
- Ensure DEVTO_API_KEY and GOOGLE_API_KEY are valid; missing keys will cause API failures.
- If you see connection errors to Dev.to, verify DEVTO_BASE_URL accessibility from your network.
- For development, you can run SSE and STDIO MCP servers in parallel; use separate terminal sessions to avoid port conflicts.
- The system relies on the DevTo service wrapper; if Dev.to API changes, update devto_service.py accordingly.
- When testing, you can call helper endpoints via the STDIO MCP server to simulate messages and verify agent responses.
- If using Windows, use proper path separators in commands or run within a POSIX-compatible shell.
Related MCP Servers
awesome-a2a
Agent2Agent (A2A) – awesome A2A agents, tools, servers & clients, all in one place.
a2a-directory
Agent2Agent (A2A) – AgentCards, Servers, Clients, Docs
awesome-a2a
Agent2Agent (A2A) – awesome A2A agents, tools, servers & clients, all in one place.
agentql
Model Context Protocol server that integrates AgentQL's data extraction capabilities.
ummon
The semantic layer for software engineering: Connect code to meaning, build on understanding
Handler
A2A Protocol client and developer toolkit.