[RFC / Discussion] Exploring standard MCP / AI-native interfaces for personal relationship context

Summary / Context

Monica has long been the gold standard for open-source personal CRM, showing the world how to thoughtfully manage interpersonal relationships, notes, and commitments with user sovereignty.

With the rapid emergence of local AI agents and the Model Context Protocol (MCP), personal relationship data is becoming the core foundation for contextual AI assistants (understanding user relationships, commitments, and conversation history).

Proposal: Connecting Monica with the Modern Agent Ecosystem

We are working on Yance, a local-first open-source communication and relationship intelligence project. We are reaching out with deep humility and respect for the years of groundwork Monica has laid.

Instead of building competing silos, we believe the community would benefit tremendously if existing, battle-tested personal CRMs like Monica could easily interoperate with modern AI agent workflows through open standards (e.g., standard MCP servers or export contracts).

Topics for Community Discussion:

1. Standardized Context Schemas: Defining lightweight schemas for contacts, relationship depth, and past interaction memories so agents can query them read-only without compromising database integrity.
2. Zero-Invasive Integrations: Allowing external open-source local agents to interact with Monica instances via read-only MCP connectors without burdening core maintainers with AI-specific features.
3. Open Specifications: We are happy to do the heavy lifting—drafting Markdown specifications, building open MCP bridge adapters, or contributing test fixtures.

We would love to hear thoughts from the maintainers and community on how Monica views interoperability with local-first AI agent ecosystems!