Skip to content
Development
Command

/memory-recall

Search and retrieve specific memories from the semantic tree based on topics, patterns, or timeframes.

From plugin
claude-command-suite
1.3k199 skills89 agents199 commands
Install
$ npx -y skills add qdhenry/Claude-Command-Suite --agent claude-code

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/memory-recall

Context preview

What this command does when you run it.

Search and retrieve specific memories from the semantic tree based on topics, patterns, or timeframes.

Command definition

memory-recall.md
tools:
  - read
  - grep
  - bash
arguments: $QUERY

Memory Recall

Search and retrieve specific memories from the semantic tree based on topics, patterns, or timeframes.

Based on the WFGY project: https://github.com/onestardao/WFGY

Instructions

1. **Parse Recall Query**

  • Extract search terms from "$QUERY"
  • Identify query type:
  • Topic search: Keywords/concepts
  • Pattern search: Logic patterns
  • Temporal search: Time ranges
  • Module search: BBMC/BBPF/BBCR/BBAM
  • Metric search: ΔS ranges

2. **Search Semantic Tree**

  • Load active tree and indices
  • Apply search filters:
  • Text matching in topics/insights
  • ΔS value ranges
  • λ_observe patterns
  • Module attribution
  • Time windows
  • Rank results by relevance:
  • Exact matches first
  • Semantic similarity
  • Recency weighting
  • Importance scores

3. **Reconstruct Context**

  • For each recalled node:
  • Retrieve full node data
  • Load surrounding context
  • Find related nodes
  • Rebuild reasoning chain
  • Calculate recall metrics:
  • Relevance score
  • Context completeness
  • Confidence level

4. **Synthesize Memories**

  • Group related recalls
  • Identify patterns
  • Extract key insights
  • Build narrative summary
  • Highlight connections

5. **Generate Recall Report**

  • Matching nodes list
  • Context reconstruction
  • Insight synthesis
  • Pattern analysis
  • Navigation suggestions

Output Format

MEMORY RECALL RESULTS
═══════════════════════════════════════
Query: "$QUERY"
Matches Found: [count]
Time Range: [earliest] to [latest]

Top Recalled Memories:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

1. [HIGH RELEVANCE] Node [id]
   Timestamp: [ISO_8601]
   Topic: "[topic matching query]"
   Module: [BBMC/BBPF/BBCR/BBAM]
   ΔS: [value] | λ: [direction]
   
   Insight: "[Full insight text with **highlighted** matches]"
   
   Context: [Previous topic] → THIS → [Next topic]
   Relevance: 95%

2. [MEDIUM RELEVANCE] Node [id]
   Timestamp: [ISO_8601]
   Topic: "[related topic]"
   Module: [module]
   ΔS: [value] | λ: [direction]
   
   Insight: "[Insight with **partial** match]"
   
   Context: [Previous] → THIS → [Next]
   Relevance: 72%

3. [PATTERN MATCH] Node [id]
   Timestamp: [ISO_8601]
   Topic: "[topic with similar pattern]"
   Module: [module]
   ΔS: [value] | λ: [direction]
   
   Insight: "[Similar reasoning pattern]"
   
   Context: [Previous] → THIS → [Next]
   Relevance: 68%

Pattern Analysis:
────────────────────────────────────────
Common Patterns Found:
• [Pattern 1]: Appears in [N] memories
• [Pattern 2]: Recurring theme
• [Pattern 3]: Similar logic flow

Temporal Distribution:
[Earliest] ████████████ [Latest]
           ↑   ↑↑  ↑
        Clusters of activity

Module Distribution:
• BBMC: [count] memories
• BBPF: [count] memories
• BBCR: [count] memories
• BBAM: [count] memories

Synthesized Insights:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Key Finding 1:
[Synthesis across multiple memories]

Key Finding 2:
[Common thread identified]

Key Finding 3:
[Evolution of understanding]

Related Topics Discovered:
• [Related topic 1] - [N] connections
• [Related topic 2] - [N] connections
• [Related topic 3] - [N] connections

Memory Clusters:
┌─────────────────────────────────────┐
│ Cluster    │ Nodes │ Theme          │
├─────────────────────────────────────┤
│ Cluster A  │ 5     │ [Theme]        │
│ Cluster B  │ 3     │ [Theme]        │
│ Cluster C  │ 2     │ [Theme]        │
└─────────────────────────────────────┘

Navigation Suggestions:
• Review full context: /semantic:tree-view --focus "[node_id]"
• Explore cluster: /memory:recall "[cluster_theme]"
• Build on insight: /semantic:node-build "[continuation]"
• Find bridges: /boundary:safe-bridge "[related_topic]"

Recall Confidence: [percentage]%
Coverage: [percentage]% of relevant memories

Recall Modes

Topic Recall

/memory:recall "machine learning"

Pattern Recall

/memory:recall --pattern "divergent-convergent"

Temporal Recall

/memory:recall --from "2024-01-01" --to "2024-01-31"

Module Recall

/memory:recall --module BBPF

Metric Recall

/memory:recall --deltaS 0.7 1.0

Advanced Search

Combined Filters

/memory:recall "API" --module BBMC --deltaS 0.5 0.8

Fuzzy Matching

/memory:recall "approximate~" --fuzzy 0.8

Regex Patterns

/memory:recall --regex "auth.*token"

Semantic Search

/memory:recall --semantic "similar to authentication"

Memory Reconstruction

The system reconstructs: 1. **Direct Memories**: Exact matches 2. **Associated Memories**: Related nodes 3. **Contextual Memories**: Surrounding nodes 4. **Pattern Memories**: Similar structures 5. **Bridge Memories**: Connecting concepts

Integration

Memory recall supports:

  • `/semantic:tree-view` for context
  • `/semantic:node-build` to continue
  • `/reasoning:chain-validate` for logic
  • `/boundary:safe-bridge` for connections
Read more
Ships withclaude-command-suite

A comprehensive development toolkit designed following Anthropic's Claude Code Best Practices for AI-assisted software development.

Get the whole plugin