Hydrate

Load context from the Obsidian memory system for session continuity.

Workflow

1. Read the living context documents:

   • Areas/AI/Context/Current State.md - Current situation, priorities, blockers
   • Areas/AI/Context/Decision Register.md - Pending decisions

2. If $ARGUMENTS provided (topic/project), search for relevant memories:

   Grep pattern="$TOPIC" path="<vault>/Areas/AI/Memory" glob="*.md"
   

   Read top 5-10 matching memory files by importance

3. If no arguments, get recent memories:

   Glob pattern="Areas/AI/Memory/**/*.md" path="<vault>"
   

   Read 5-10 most recent memory files

4. Scan recent session logs for continuity:

   Glob pattern="Areas/AI/Collaboration/Sessions/*.md" path="<vault>"
   

   Read 2-3 most recent sessions

5. Internalize the context silently - don't recite it back verbatim

6. Acknowledge briefly:

   • Key context points from Current State
   • Number of memories loaded
   • Recent session topics
   • Ready to proceed

Context Sources

Examples

User: /hydrate -> Load Current State, recent memories, recent sessions -> "Loaded context: Current State shows interview prep priority. 8 recent memories loaded. Last session covered statusline config. Ready."

User: /hydrate metatron -> Load Current State + search memories for "metatron" -> "Loaded context: Found 5 memories about Metatron (deployment, stack decisions, design). Current priority is interview prep. Ready."

User: /hydrate robabby -> Load Current State + search memories for "robabby" -> "Loaded context: Found 7 memories about robabby.com (sacred geometry redesign, PRs, patterns learned). Ready."

Response Format

Brief acknowledgment:

• Context loaded (Current State key points if relevant)
• Number of memories found
• Recent session topics if applicable
• "Ready to proceed" or similar