Configure AI agents for automated content curation with Curyloop.
AI Agent Setup
Curyloop works with any MCP-compatible AI client. Here are the most popular setups with ready-to-use prompts.
Claude Code
bash
claude mcp add curyloop \
--transport http \
--url https://curyloop.com/api/mcp \
--header "Authorization: Bearer YOUR_API_KEY"
Prompt:
Search the web for the best resources about "edge computing in 2026",
then add the top 5 results to my Curyloop session called "Tech Trends".
Use high priority for articles from major publications.
Cline (VS Code)
Add to your .vscode/mcp.json:
json
{
"mcpServers": {
"curyloop": {
"type": "streamable-http",
"url": "https://curyloop.com/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}
Prompt:
Look at the packages in my project's package.json, find the official
documentation and best tutorial for each major dependency, and add
them to my "Project Resources" session in Curyloop.
Cursor
Add to your .cursor/mcp.json:
json
{
"mcpServers": {
"curyloop": {
"type": "streamable-http",
"url": "https://curyloop.com/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}
Prompt:
Research the best practices for the tech stack in this project,
find 5 high-quality guides, and save them to my "Best Practices"
session in Curyloop with appropriate priorities.
Get your API key from Settings > API Keys. For detailed MCP setup, see MCP Setup.