Connect any MCP-compatible AI tool to the Relational Tech Project commons — the Studio's remixable builder tools, stories, and prompts, plus methodology, neighborhood recipes, frameworks, and field references. Search is semantic: describe a need in plain language.
Add this to your AI tool's MCP configuration:
{
"mcpServers": {
"relational-tech": {
"type": "streamable-http",
"url": "https://mcp.relationaltechproject.org/mcp"
}
}
}
Endpoint:
https://mcp.relationaltechproject.org/mcp
.mcp.json or run claude mcp add relational-tech --type streamable-http --url https://mcp.relationaltechproject.org/mcpNo API key. No signup. Reads are public via row-level security; contributions land in a stewarded review queue — nothing publishes without a human.
Once connected, invoke the practice-guide prompt to put your AI in a Neighboring Commons practice-guide stance for the rest of the conversation: relationships-first, asset-based, citing practitioners by name, calibrating confidence honestly.
Your AI will surface relevant recipes, frameworks, methodology, and practitioner references — with attribution and source URLs so you can read the original work.
6 tools: search-studio-library, get-tool-details, find-patterns-by-context, suggest-contribution, submit-contribution, get-network-updates
6 prompts: practice-guide, design-neighborhood-tool, assess-relational-soil, create-builder-action-plan, remix-existing-tool, create-build-plan
9 resources at rtp://knowledge/* URIs (methodology docs, queried live from the commons)
Open source on GitHub. See the README for the full architecture, deployment guide, and how the commons works.
Health check: /health