[Memory P1] Seed holographic fact store from existing knowledge #240

Open
opened 2026-04-09 00:41:06 +00:00 by Rockachopa · 0 comments
Owner

Context

The holographic fact store (memory_store.db) has full prefetch wiring — it searches on every user message and injects matching facts. But it contains 0 facts. The engine is running but the tank is empty.

Task

Bootstrap the fact store with existing knowledge:

  1. Extract all entries from MEMORY.md and USER.md into structured facts with proper categories
  2. Add facts from skills knowledge (wizard roster, repo purposes, fleet architecture)
  3. Add key facts from SOUL.md (behavioral guidelines, core values)
  4. Tag entities properly (Alexander, Ezra, Bezalel, Allegro, each repo name)
  5. Set appropriate trust scores (user-stated facts = 0.9, inferred = 0.6)

Acceptance Criteria

  • fact_store(action='list') returns 50+ facts
  • Facts have proper entity tags (probe("Ezra") returns Ezra facts)
  • Facts have appropriate categories (user_pref, project, tool, general)
  • prefetch("tell me about Ezra") returns relevant facts

Notes

This is a one-time bootstrap. Ongoing fact accumulation comes from session-end extraction and agent tool calls.

Part of: [EPIC] Unified Memory Architecture

## Context The holographic fact store (memory_store.db) has full prefetch wiring — it searches on every user message and injects matching facts. But it contains 0 facts. The engine is running but the tank is empty. ## Task Bootstrap the fact store with existing knowledge: 1. Extract all entries from MEMORY.md and USER.md into structured facts with proper categories 2. Add facts from skills knowledge (wizard roster, repo purposes, fleet architecture) 3. Add key facts from SOUL.md (behavioral guidelines, core values) 4. Tag entities properly (Alexander, Ezra, Bezalel, Allegro, each repo name) 5. Set appropriate trust scores (user-stated facts = 0.9, inferred = 0.6) ## Acceptance Criteria - `fact_store(action='list')` returns 50+ facts - Facts have proper entity tags (probe("Ezra") returns Ezra facts) - Facts have appropriate categories (user_pref, project, tool, general) - prefetch("tell me about Ezra") returns relevant facts ## Notes This is a one-time bootstrap. Ongoing fact accumulation comes from session-end extraction and agent tool calls. Part of: [EPIC] Unified Memory Architecture
Rockachopa added this to the Unified Memory Architecture milestone 2026-04-09 00:41:06 +00:00
Rockachopa added the phase:1-activateepic:memory labels 2026-04-09 00:41:06 +00:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/hermes-agent#240