varun29ankus avatar

Mif Tools

varun29ankus/mif-spec

If you're tired of losing your AI agent's memory every time you switch systems, this MCP server wraps the Memory Interchange Format spec into five tools: export_memories, import_memories, validate_memories, inspect_memories, and list_formats. It auto-detects and converts between mem0, CrewAI, LangChain, generic JSON, and Markdown formats using a minimal three-field schema (id, content, created_at) with optional support for embeddings, entity references, and knowledge graphs. The underlying mif-tools library has zero dependencies for core functionality and preserves vendor extensions on round-trip. Think of it as vCard for agent memories. Install with pip install mif-tools[mcp] or grab the npm package for TypeScript projects.

tools
5
transport
STDIO
last updated
Mar 6, 2026

Tools

5 tools

Public tool metadata for what this MCP can expose to an agent.

export_memoriesExport memories from a storage backend into MIF format.

Export memories from a storage backend into MIF format.

No parameter schema in public metadata yet.

import_memoriesImport memories from any supported format into the storage backend.

Import memories from any supported format into the storage backend.

No parameter schema in public metadata yet.

validate_memories

No parameter schema in public metadata yet.

inspect_memories

No parameter schema in public metadata yet.

list_formats

No parameter schema in public metadata yet.

Sponsors

Readme

Memory Interchange Format (MIF)

PyPI npm License Tests Docs

Your AI agent has 6 months of memories in System A. You want to try System B. Without MIF, you lose everything. With MIF:

pip install mif-tools
mif convert mem0_export.json --to shodh -o memories.mif.json

Done. Your memories are portable.

What is MIF?


View the full README on GitHub

Details

Repository
github.com/varun29ankus/mif-spec
Category
AI & LLM Tools
Package
mif-tools
Transport
STDIO
Registry
MCP registry, active

More from varun29ankus

Related AI & LLM Tools MCP servers

View all
1jigyasudham avatarVetojigyasudham/veto93 agentic MCP tools + 49 specialist agents for every major AI CLI. Self-learning, no API keys.4stars
2jmcentire avatarsignet-evaljmcentire/signet-evalDeterministic policy enforcement and MCP management for AI agent tool calls.4stars
3joinwell52-ai avatarFcopjoinwell52-ai/fcopFCoP Core adapter with 25 canonical tools and 6 read-only resources for durable agent work.4stars
4penfieldlabs avatarPenfield Memorypenfieldlabs/penfield-mcpPersistent memory and knowledge graphs for AI agents. Hybrid search, context checkpoints, and more.4stars
5tessera-llm avatarTessera MCP Servertessera-llm/mcp-serverMCP server for Tessera's LLM cost-optimization layer: anchored spend, drift, recommendations.0stars
6agentled avatarMcp Serveragentled/mcp-serverIntelligent AI workflow orchestration with long-term memory, 100+ integrations, and unified credits.0stars