Memory Map

Render the soul's memory structure visually. Shows attractors (conceptual gravity wells), their basins, session context, and optionally the neighborhood around a specific query.

Execute directly — Produces formatted output for terminal and documentation.

How to Execute

Default Mode (attractors + session)

# Step 1: Get soul state and attractors
chitta soul_context format="json"
chitta attractors max_attractors=7

# Step 2: Render ASCII summary
=== Soul Memory Map ===

Coherence: XX% [████████░░]  Ojas: XX% [█████████░]
Nodes: XXXX total (XXXX hot, XX warm, XX cold)

ATTRACTORS (X gravity wells)
  [1] first 40 chars of label.........(basin: XX)
  [2] second attractor label..........(basin: XX)
  ...

SESSION
  Primed: XX recent observations
  Active: X intentions
  Basin: XX goal-adjacent nodes

# Step 3: Render Mermaid diagram

graph TD
    subgraph "Attractor: label"
        A1((strength%))
    end
    N1[basin node] --> A1
    N2[basin node] --> A1

Attractors Mode (/memory-map attractors)

Focus on attractor basins with more detail:

# Get attractors with higher limit
chitta attractors max_attractors=10

# Render detailed basin view
=== Attractor Basins ===

[1] ATTRACTOR: label (strength: XX%)
    Basin size: XX nodes
    ├─ Node in basin
    ├─ Another node
    └─ Third node

[2] ATTRACTOR: label (strength: XX%)
    ...

Query Mode (/memory-map query <text>)

# Step 1: Search with dense zoom to get edges
chitta recall query="<text>", zoom="dense", limit=8

# Step 2: Get soul state for context
chitta soul_context format="json"

# Step 3: Render neighborhood
=== Memory Neighborhood: "<query>" ===

[87%] Node title (wisdom)
      └─Similar→ Related node (0.8)
      └─AppliedIn→ Another node (0.6)
[72%] Second result (episode)
      └─Supports→ Supporting node (0.9)
...

# Step 4: Render Mermaid showing edges

graph LR
    Q[Query] -.87%.-> N1[Node title]
    N1 --Similar--> N2[Related node]
    N1 --AppliedIn--> N3[Another node]

Session Mode (/memory-map session)

# Get full session context with ledger
chitta soul_context format="json", include_ledger=true

# Render session state
=== Session Context ===

PRIMING
  Recent observations: XX nodes
  Active intentions: X nodes
  Goal basin: XX nodes

COMPETITION
  Enabled: yes/no
  Threshold: XX%
  Inhibition: XX%

LEDGER (if present)
  Work: X todos
  Next: X steps
  Critical: X items

Output Formats

ASCII Progress Bars

Build 10-char progress bars from percentage:

0-10%: [░░░░░░░░░░]
50%: [█████░░░░░]
100%: [██████████]

Formula: filled = round(percentage / 10), empty = 10 - filled

Edge Type Symbols

Edge Type	ASCII Symbol	Mermaid Arrow
Similar	`└─Similar→`	`--Similar-->`
Supports	`└─Supports→`	`--Supports-->`
Contradicts	`└─Contradicts→`	`-.Contradicts.->`
AppliedIn	`└─AppliedIn→`	`--AppliedIn-->`
EvolvedFrom	`└─EvolvedFrom→`	`--EvolvedFrom-->`
PartOf	`└─PartOf→`	`--PartOf-->`
TriggeredBy	`└─TriggeredBy→`	`--TriggeredBy-->`

Node Type Labels

Show type in parentheses after title:

(wisdom) - Durable knowledge
(belief) - Core convictions
(episode) - Episodic observations
(intention) - Active goals
(failure) - Learned mistakes
(dream) - Aspirational ideas

When to Use

Orientation: Start of session, understand what's in memory
Debugging: Why did this memory surface? Check edges and relevance
Planning: What attractors exist? Where should new knowledge go?
Health check: Are there orphan nodes? Disconnected clusters?
Documentation: Export Mermaid diagrams for architecture docs

memory-map