Skip to content

Personal CRM with AI-managed contact profiles and per-person tone matching #704

@kovtcharov

Description

@kovtcharov

Summary

AI-managed contact relationship management that learns from your communication patterns. The agent builds a profile for each contact — communication history, preferred channels, tone, relationship context — and uses it to draft messages that match how you actually communicate with each person.

Key Features

  1. Auto-built contact profiles — aggregated from email, SMS, messaging interactions
  2. Per-person tone matching — agent drafts responses in exactly how you communicate with each specific person (formal with boss, casual with friends)
  3. Cross-channel profile linking — same person across email, text, LinkedIn, messaging
  4. Communication pattern tracking — frequency, response time, preferred channel per contact
  5. Relationship context — agent remembers what you've discussed with each person

Privacy

Contact profiles and communication patterns are among the most sensitive personal data. This is a category where local processing is not just preferred — it's the only acceptable architecture. No cloud service should have access to your full relationship graph.

Dependencies

Acceptance Criteria

  • Contact profiles auto-created from communication history
  • Tone profiles generated per contact (verified by user)
  • Draft responses match per-contact communication style
  • Cross-channel linking (same person across platforms)
  • All profile data stored locally in SQLite

Metadata

Metadata

Assignees

No one assigned

    Labels

    agentdomain:automationScheduler, autonomy, RAG, web search, watchers, researchenhancementNew feature or requestp3track:consumer-appHermes-competitor consumer product — mobile-first, voice + messaging + memory + skills

    Type

    No type

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions