Skip to content

deepadata/deepadata-adapters

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

deepadata-adapters

EDM enrichment adapters for memory platforms.

Memory platforms store what was said. EDM captures the emotional context of why it mattered — as a governed, portable artifact.

Packages

Package Description npm
@deepadata/mem0-adapter EDM enrichment for Mem0 npm install @deepadata/mem0-adapter
@deepadata/zep-adapter EDM enrichment for Zep npm install @deepadata/zep-adapter
@deepadata/langchain-adapter EDM enrichment for LangChain npm install @deepadata/langchain-adapter

Architecture

All adapters follow the enrichment layer pattern:

// EDM enrichment runs ALONGSIDE your memory platform, not instead of it
const { edmArtifact } = await enrichWithEDM(text, { profile: 'essential' });

// Your memory platform continues unchanged
await mem0.add(text, { user_id: 'user123' });
await zep.memory.add(sessionId, { messages });
await memory.saveContext({ input }, { output });

Profiles

Profile Fields Use Case
essential ~20 Memory platforms, agent frameworks (default for Mem0, LangChain)
extended ~45 Journaling, companion AI, workplace wellness (default for Zep)
full 96 Clinical, therapeutic, regulated systems

Provider Selection

All adapters default to Kimi K2 for cost optimization. For GDPR Article 9 or HIPAA regulated deployments, specify anthropic or openai:

const { edmArtifact } = await enrichWithEDM(text, {
  profile: 'essential',
  provider: 'anthropic',  // EU/US compliance
});

Development

# Install dependencies
npm install

# Build all packages
npm run build

# Typecheck all packages
npm run typecheck

Related

License

MIT

About

EDM enrichment layer adapters for Mem0, Zep, and LangChain — capture emotional context alongside memory storage

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors