Skill This
AI Skill Report Card

Memory Read & Access Operations

B+75·Feb 13, 2026·Source: Extension-page

Memory - Read & Access Operations

Unified read-side memory skill with subcommands for searching, loading, syncing, history, and visualization.

Quick Start15 / 15
Bash
/ork:memory search "authentication patterns"     # Search decisions
/ork:memory load                                # Load session context  
/ork:memory viz --entity "database"            # Visualize relationships
/ork:memory history --since 7d                 # Recent decisions
Recommendation▾
Simplify the workflow section - the Progress checklist format is unnecessary overhead for a read operation
Workflow12 / 15

1. Parse Command Structure

Progress:
- [ ] Extract subcommand from first argument
- [ ] Parse remaining flags and options
- [ ] Validate subcommand exists
- [ ] Route to appropriate handler

If no subcommand provided:

Python
AskUserQuestion(
    questions=[{
        "question": "What memory operation do you need?",
        "header": "Operation", 
        "options": [
            {"label": "search", "description": "Search decisions and patterns in knowledge graph"},
            {"label": "load", "description": "Load relevant context for this session"},
            {"label": "sync", "description": "Sync decisions to mem0 cloud"},
            {"label": "history", "description": "View decision timeline"},
            {"label": "viz", "description": "Visualize knowledge graph as Mermaid"},
            {"label": "status", "description": "Check memory system health"}
        ],
        "multiSelect": false
    }]
)

2. Execute Subcommands

search - Search Knowledge Graph

Bash
/ork:memory search <query>                    # Basic search
/ork:memory search --category <cat> <query>   # Filter by category  
/ork:memory search --mem0 <query>            # Include cloud search
/ork:memory search --global <query>          # Cross-project patterns

Search workflow:

  1. Parse flags (--category, --limit, --mem0, --agent, --global)
  2. Build filters from flags
  3. Search graph via mcp__memory__search_nodes
  4. If --mem0 flag: parallel search mem0 cloud
  5. Merge and deduplicate results

load - Load Session Context

Bash
/ork:memory load                    # Load all relevant context
/ork:memory load --project          # Project-specific only
/ork:memory load --global           # Include global best practices

sync - Sync to Mem0 Cloud

Bash
/ork:memory sync                    # Sync pending changes
/ork:memory sync --force            # Force full sync
/ork:memory sync --dry-run          # Preview sync

history - Decision Timeline

Bash
/ork:memory history                 # Recent decisions
/ork:memory history --mermaid       # Timeline diagram
/ork:memory history --since 7d      # Last 7 days

viz - Knowledge Graph Visualization

Bash
/ork:memory viz                     # Full graph
/ork:memory viz --entity <name>     # Focus entity
/ork:memory viz --depth 2           # Limit depth

status - System Health

Bash
/ork:memory status                  # Show system status
Recommendation▾
Reduce redundancy between subcommand descriptions in workflow and examples sections
Examples15 / 20

Example 1: Search for Authentication Patterns Input: /ork:memory search "JWT authentication" Output:

Found 3 results matching "JWT authentication":

[GRAPH] jwt-pattern (Pattern)
-> USES -> express-jwt-middleware
-> ENABLES -> stateless-auth
Observations: Preferred for API authentication, expires in 24h

[GRAPH] auth-decision-2023-11 (Decision)  
-> CHOSE_OVER -> session-cookies
-> REQUIRES -> secret-rotation

Example 2: Load Project Context Input: /ork:memory load --project Output:

Loaded session context:
- 5 recent decisions from current project
- 12 active patterns (database, auth, caching)
- 3 agent-specific memories
- 2 pending implementation tasks

Example 3: Visualize Database Relationships Input: /ork:memory viz --entity "postgresql" Output:

MERMAID
graph TD
    postgresql[PostgreSQL]
    postgresql --> pgvector[pgvector extension]
    postgresql --> connection_pool[Connection Pool]
    pgvector --> vector_search[Vector Search]
    connection_pool --> performance[Performance]
Recommendation▾
Add more concrete input/output pairs showing actual graph data structures and search results
Best Practices
  • Use --global flag to access cross-project patterns
  • Combine --mem0 with local search for comprehensive results
  • Load context at session start for continuity
  • Use specific entity names in visualization for focused views
  • Sync regularly to preserve decisions across sessions
  • Check status if experiencing memory issues
Common Pitfalls
  • Don't search without context - use category filters for large graphs
  • Don't skip loading context - it provides session continuity
  • Don't forget to sync before ending sessions
  • Don't visualize entire large graphs - use depth/entity limits
  • Don't ignore status warnings - corrupt memory affects performance
0
Grade B+AI Skill Framework
Scorecard
Criteria Breakdown
Quick Start
15/15
Workflow
12/15
Examples
15/20
Completeness
15/20
Format
15/15
Conciseness
13/15
Save to My Skills