@mindrian_os/cli 1.13.0-beta.10

package/references/brain/causal-enrichment.cypher

@@ -0,0 +1,165 @@
+// Brain Causal Framework Enrichment
+// Phase 52: BRAIN-01 through BRAIN-06
+// Execute via mcp__neo4j-brain__write_neo4j_cypher
+//
+// IMPORTANT: Before running any write statements, run the discovery queries
+// in the "Discovery" section to verify exact node names. Replace placeholder
+// names below with exact names from discovery results.
+//
+// Generated: 2026-04-05
+// Requirements: BRAIN-01, BRAIN-02, BRAIN-03, BRAIN-04, BRAIN-05, BRAIN-06
+
+// ============================================================================
+// SECTION 0: DISCOVERY (run FIRST via mcp__neo4j-brain__read_neo4j_cypher)
+// ============================================================================
+
+// Query A: Get all Framework node names (verify exact spelling)
+// MATCH (f:Framework) RETURN f.name AS name LIMIT 50
+
+// Query B: Get all Concept node names
+// MATCH (c:Concept) RETURN c.name AS name LIMIT 50
+
+// Query C: Check VentureStage nodes
+// MATCH (v:VentureStage) RETURN v.name AS name LIMIT 20
+
+// Query D: Check existing edge types on Framework nodes
+// MATCH (f:Framework)-[r]->() RETURN DISTINCT type(r) AS rel_type LIMIT 30
+
+// Query E: Check if Falsifiability or Hypothesis Tree exist
+// MATCH (f:Framework) WHERE f.name CONTAINS 'Falsif' OR f.name CONTAINS 'Hypothesis' RETURN f.name LIMIT 10
+
+// ============================================================================
+// SECTION 1: BRAIN-03 -- Create Theory of Change Framework node
+// ============================================================================
+
+MERGE (f:Framework {name: 'Theory of Change'})
+ON CREATE SET f.description = 'Forward causal reasoning: articulate the causal pathway from activities to outcomes. Maps assumptions about how change happens.',
+              f.category = 'causal',
+              f.created = datetime()
+
+// ============================================================================
+// SECTION 2: BRAIN-04 -- Create Causal Reasoning parent Concept node
+// ============================================================================
+
+MERGE (c:Concept {name: 'Causal Reasoning'})
+ON CREATE SET c.description = 'Parent concept connecting causal analysis frameworks: root cause analysis, systems thinking, causal loop diagrams, scenario analysis, and theory of change.',
+              c.created = datetime()
+
+// Link each causal framework to the Causal Reasoning concept
+// NOTE: Use exact names from Query A. Names below are best-guess from Brain schema.
+
+MATCH (f:Framework {name: 'Root Cause Analysis'}), (c:Concept {name: 'Causal Reasoning'})
+MERGE (f)-[:RELATED_TO]->(c)
+
+MATCH (f:Framework {name: 'Systems Thinking'}), (c:Concept {name: 'Causal Reasoning'})
+MERGE (f)-[:RELATED_TO]->(c)
+
+MATCH (f:Framework {name: 'Causal Loop Diagrams'}), (c:Concept {name: 'Causal Reasoning'})
+MERGE (f)-[:RELATED_TO]->(c)
+
+MATCH (f:Framework {name: 'Scenario Analysis Framework'}), (c:Concept {name: 'Causal Reasoning'})
+MERGE (f)-[:RELATED_TO]->(c)
+
+MATCH (f:Framework {name: 'Theory of Change'}), (c:Concept {name: 'Causal Reasoning'})
+MERGE (f)-[:RELATED_TO]->(c)
+
+// ============================================================================
+// SECTION 3: BRAIN-01 -- Wire FEEDS_INTO chains
+// ============================================================================
+
+// Chain: Root Cause Analysis -> Systems Thinking -> Causal Loop Diagrams -> Scenario Analysis
+MATCH (a:Framework {name: 'Root Cause Analysis'}), (b:Framework {name: 'Systems Thinking'})
+MERGE (a)-[:FEEDS_INTO]->(b)
+
+MATCH (a:Framework {name: 'Systems Thinking'}), (b:Framework {name: 'Causal Loop Diagrams'})
+MERGE (a)-[:FEEDS_INTO]->(b)
+
+MATCH (a:Framework {name: 'Causal Loop Diagrams'}), (b:Framework {name: 'Scenario Analysis Framework'})
+MERGE (a)-[:FEEDS_INTO]->(b)
+
+// Branch: Systems Thinking -> Reverse Salient Analysis
+MATCH (a:Framework {name: 'Systems Thinking'}), (b:Framework {name: 'Reverse Salient Analysis'})
+MERGE (a)-[:FEEDS_INTO]->(b)
+
+// ============================================================================
+// SECTION 4: BRAIN-02 -- CO_OCCURS edges (bidirectional)
+// ============================================================================
+
+// Root Cause Analysis <-> Six Thinking Hats
+MATCH (a:Framework {name: 'Root Cause Analysis'}), (b:Framework {name: 'Six Thinking Hats'})
+MERGE (a)-[:CO_OCCURS]->(b)
+MERGE (b)-[:CO_OCCURS]->(a)
+
+// Systems Thinking <-> Reverse Salient Analysis
+MATCH (a:Framework {name: 'Systems Thinking'}), (b:Framework {name: 'Reverse Salient Analysis'})
+MERGE (a)-[:CO_OCCURS]->(b)
+MERGE (b)-[:CO_OCCURS]->(a)
+
+// Cynefin <-> Root Cause Analysis
+MATCH (a:Framework {name: 'Cynefin Framework'}), (b:Framework {name: 'Root Cause Analysis'})
+MERGE (a)-[:CO_OCCURS]->(b)
+MERGE (b)-[:CO_OCCURS]->(a)
+
+// ============================================================================
+// SECTION 5: BRAIN-05 -- TYPICAL_AT venture stage mappings
+// ============================================================================
+
+// Create VentureStage nodes if they do not exist (verify with Query C first)
+MERGE (s:VentureStage {name: 'Pre-Opportunity'})
+MERGE (s2:VentureStage {name: 'Opportunity Identified'})
+MERGE (s3:VentureStage {name: 'Problem Validation'})
+
+// Map causal frameworks to venture stages
+MATCH (f:Framework {name: 'Root Cause Analysis'}), (s:VentureStage {name: 'Opportunity Identified'})
+MERGE (f)-[:TYPICAL_AT]->(s)
+
+MATCH (f:Framework {name: 'Systems Thinking'}), (s:VentureStage {name: 'Pre-Opportunity'})
+MERGE (f)-[:TYPICAL_AT]->(s)
+
+MATCH (f:Framework {name: 'Reverse Salient Analysis'}), (s:VentureStage {name: 'Pre-Opportunity'})
+MERGE (f)-[:TYPICAL_AT]->(s)
+
+MATCH (f:Framework {name: 'Theory of Change'}), (s:VentureStage {name: 'Problem Validation'})
+MERGE (f)-[:TYPICAL_AT]->(s)
+
+// ============================================================================
+// SECTION 6: BRAIN-06 -- Link Falsifiability and Hypothesis Tree
+// ============================================================================
+
+// NOTE: Use exact names from Query E. Names below are best-guess.
+
+// Link to Causal Reasoning concept
+// NOTE: Falsifiability is a Concept, Logic Trees is a Concept (not Framework labels)
+MATCH (f:Concept {name: 'Falsifiability'}), (c:Concept {name: 'Causal Reasoning'})
+MERGE (f)-[:RELATED_TO]->(c)
+
+MATCH (f:Concept {name: 'Logic Trees (Issue, Hypothesis, Decision)'}), (c:Concept {name: 'Causal Reasoning'})
+MERGE (f)-[:RELATED_TO]->(c)
+
+// FEEDS_INTO from Theory of Change to Logic Trees
+MATCH (a:Framework {name: 'Theory of Change'}), (b:Concept {name: 'Logic Trees (Issue, Hypothesis, Decision)'})
+MERGE (a)-[:FEEDS_INTO]->(b)
+
+// FEEDS_INTO from Logic Trees to Falsifiability
+MATCH (a:Concept {name: 'Logic Trees (Issue, Hypothesis, Decision)'}), (b:Concept {name: 'Falsifiability'})
+MERGE (a)-[:FEEDS_INTO]->(b)
+
+// ============================================================================
+// VERIFICATION QUERIES (run via mcp__neo4j-brain__read_neo4j_cypher)
+// ============================================================================
+
+// Verify BRAIN-01: FEEDS_INTO chain
+// MATCH (a:Framework)-[:FEEDS_INTO]->(b:Framework)
+// WHERE a.name CONTAINS 'Root Cause' OR a.name CONTAINS 'Systems' OR a.name CONTAINS 'Causal Loop'
+// RETURN a.name AS from_fw, b.name AS to_fw
+// LIMIT 10
+
+// Verify BRAIN-04: Causal Reasoning concept connections
+// MATCH (f:Framework)-[:RELATED_TO]->(c:Concept {name: 'Causal Reasoning'})
+// RETURN f.name AS framework
+// LIMIT 10
+
+// Verify BRAIN-05: TYPICAL_AT edges
+// MATCH (f:Framework)-[:TYPICAL_AT]->(s:VentureStage)
+// RETURN f.name AS framework, s.name AS stage
+// LIMIT 10

package/references/brain/command-triggers-schema.md

@@ -0,0 +1,226 @@
+# Brain Command Trigger Schema
+
+## Purpose
+
+Commands are first-class nodes in the Neo4j Brain. They have typed relationships to Frameworks, ProblemTypes, VentureStages, and Signals that define WHEN they should fire. The Brain doesn't just recommend frameworks -- it recommends ACTIONS mapped to specific /mos: commands with trigger conditions.
+
+## Node: Command
+
+```cypher
+CREATE (c:Command {
+  name: '/mos:find-analogies',
+  slug: 'find-analogies',
+  category: 'intelligence',          // infrastructure | methodology | intelligence | export | meeting | funding
+  description: 'Cross-domain analogy discovery via structural isomorphism',
+  jtbd_when: 'stuck on a problem that feels unique to your domain',
+  jtbd_want: 'discover that other industries already solved the same structural conflict',
+  jtbd_so: 'adapt their approach instead of inventing from scratch',
+  time_estimate: '5 minutes',
+  flags: ['--brain', '--external'],   // available flags
+  min_sections: 2,                    // minimum populated sections to be relevant
+  requires_brain: false,              // works without Brain (but better with)
+  powerhouse: true                    // v1.6.0+ feature
+})
+```
+
+## Relationships
+
+### TRIGGERED_BY_SIGNAL
+
+Links a Command to the Room Signal types that should trigger it.
+
+```cypher
+// When a CONTRADICTS edge exists between sections -> suggest find-analogies
+CREATE (cmd:Command {name: '/mos:find-analogies'})-[:TRIGGERED_BY_SIGNAL {
+  signal_type: 'TENSION',            // TENSION | BLIND_SPOT | BOTTLENECK | PATTERN | SURPRISE
+  priority: 1,                        // 1=highest, 5=lowest
+  condition: 'contradicts_count >= 1',
+  jtbd_framing: 'When your Sections have unresolved Tensions, you want to see how other domains resolved the same structural conflict'
+}]->(sig:SignalType {name: 'TENSION'})
+
+// When 3+ sections have gaps -> suggest act --swarm
+CREATE (cmd:Command {name: '/mos:act --swarm'})-[:TRIGGERED_BY_SIGNAL {
+  signal_type: 'BLIND_SPOT',
+  priority: 1,
+  condition: 'gap_count >= 3',
+  jtbd_framing: 'When 3+ Sections have Blind Spots, you want to fill them all at once instead of one at a time'
+}]->(sig:SignalType {name: 'BLIND_SPOT'})
+
+// When reverse salients detected -> suggest specific methodology for lagging section
+CREATE (cmd:Command {name: '/mos:act'})-[:TRIGGERED_BY_SIGNAL {
+  signal_type: 'BOTTLENECK',
+  priority: 2,
+  condition: 'reverse_salient_count >= 1',
+  jtbd_framing: 'When a Section lags behind the others, you want to close the gap before it compounds'
+}]->(sig:SignalType {name: 'BOTTLENECK'})
+```
+
+### FOLLOWS_FRAMEWORK
+
+Links a Command to the Framework it should follow (chaining).
+
+```cypher
+// After JTBD analysis -> suggest Blue Ocean
+CREATE (cmd:Command {name: '/mos:dominant-designs'})-[:FOLLOWS_FRAMEWORK {
+  confidence: 0.85,
+  reason: 'JTBD reveals customer needs; dominant designs reveals competitive whitespace',
+  hop_distance: 1
+}]->(f:Framework {name: 'Jobs-to-Be-Done'})
+
+// After find-analogies -> suggest validate (stress-test the analogy)
+CREATE (cmd:Command {name: '/mos:validate'})-[:FOLLOWS_FRAMEWORK {
+  confidence: 0.90,
+  reason: 'Analogies must be stress-tested before acting on them',
+  hop_distance: 1
+}]->(f:Framework {name: 'Design-by-Analogy'})
+
+// After grade --full -> suggest act --swarm on weak sections
+CREATE (cmd:Command {name: '/mos:act --swarm'})-[:FOLLOWS_FRAMEWORK {
+  confidence: 0.80,
+  reason: 'Grading reveals gaps; swarm attacks them in parallel',
+  hop_distance: 1
+}]->(f:Framework {name: 'Assessment'})
+```
+
+### RELEVANT_AT_STAGE
+
+Links a Command to VentureStages where it's most impactful.
+
+```cypher
+// find-analogies is most powerful at Discovery and Design stages
+CREATE (cmd:Command {name: '/mos:find-analogies'})-[:RELEVANT_AT_STAGE {
+  impact: 'high',
+  reason: 'Discovery needs fresh perspectives; Design needs proven patterns'
+}]->(stage:VentureStage {name: 'Discovery'})
+
+// grade --full becomes critical at Investment stage
+CREATE (cmd:Command {name: '/mos:grade --full'})-[:RELEVANT_AT_STAGE {
+  impact: 'critical',
+  reason: 'Investment stage demands rigor across every Section'
+}]->(stage:VentureStage {name: 'Investment'})
+
+// scout is always relevant but especially at Validation
+CREATE (cmd:Command {name: '/mos:scout'})-[:RELEVANT_AT_STAGE {
+  impact: 'high',
+  reason: 'Validation stage has deadlines and competitor dynamics'
+}]->(stage:VentureStage {name: 'Validation'})
+```
+
+### ADDRESSES_PROBLEM_TYPE
+
+Links a Command to ProblemTypes it solves (via multi-hop through Frameworks).
+
+```cypher
+// find-analogies addresses wicked problems (via Design-by-Analogy framework)
+CREATE (cmd:Command {name: '/mos:find-analogies'})-[:ADDRESSES_PROBLEM_TYPE {
+  via_framework: 'Design-by-Analogy',
+  effectiveness: 0.85
+}]->(pt:ProblemType {name: 'wicked'})
+
+// act --swarm addresses complex problems with multiple dimensions
+CREATE (cmd:Command {name: '/mos:act --swarm'})-[:ADDRESSES_PROBLEM_TYPE {
+  via_framework: 'Multiple simultaneous frameworks',
+  effectiveness: 0.90
+}]->(pt:ProblemType {name: 'ill-defined-complex'})
+```
+
+### RESOLVES_TENSION_BETWEEN
+
+Links a Command to Framework pairs that commonly produce contradictions.
+
+```cypher
+// find-analogies resolves tensions between pricing frameworks and competitive positioning
+CREATE (cmd:Command {name: '/mos:find-analogies'})-[:RESOLVES_TENSION_BETWEEN {
+  framework_a: 'Lean Canvas',
+  framework_b: 'Competitive Analysis',
+  resolution_pattern: 'Cross-domain structural isomorphism finds how other industries resolved pricing vs positioning'
+}]->(tension:TensionType {name: 'pricing_vs_positioning'})
+```
+
+## Multi-Hop Proactive Query
+
+The power of this schema is MULTI-HOP reasoning. The Brain can traverse:
+
+```
+User's Room State
+  -> Current Frameworks (from Room artifacts)
+    -> FEEDS_INTO relationships (what comes next)
+      -> Command nodes (which /mos: command delivers it)
+        -> TRIGGERED_BY_SIGNAL (is the trigger condition met?)
+          -> Return ranked command suggestion with full JTBD framing
+```
+
+### Query: brain_proactive_command (Pattern 10d)
+
+```cypher
+// Multi-hop: Room frameworks -> next framework -> command -> trigger check
+MATCH (current:Framework)<-[:FOLLOWS_FRAMEWORK]-(cmd:Command)
+WHERE current.name IN $room_frameworks
+
+// Check if command is relevant at this venture stage
+OPTIONAL MATCH (cmd)-[stage_rel:RELEVANT_AT_STAGE]->(stage:VentureStage {name: $venture_stage})
+
+// Check if command is triggered by current signals
+OPTIONAL MATCH (cmd)-[trigger:TRIGGERED_BY_SIGNAL]->(sig:SignalType)
+
+// Get success data
+OPTIONAL MATCH (current)-[:APPLIED_IN]->(example:Example)
+WHERE example.grade_numeric >= 80
+
+RETURN cmd.name AS command,
+       cmd.jtbd_when AS when_situation,
+       cmd.jtbd_want AS want_motivation,
+       cmd.jtbd_so AS so_outcome,
+       cmd.time_estimate AS time,
+       cmd.flags AS available_flags,
+       stage_rel.impact AS stage_impact,
+       trigger.signal_type AS triggered_by,
+       trigger.condition AS trigger_condition,
+       trigger.jtbd_framing AS trigger_framing,
+       count(example) AS success_count
+ORDER BY stage_rel.impact DESC, trigger.priority ASC, success_count DESC
+LIMIT 5
+```
+
+**Parameters:**
+- `$room_frameworks` -- frameworks used in Room (from artifact frontmatter)
+- `$venture_stage` -- from STATE.md
+
+**Output:** Ranked command suggestions with full JTBD framing, trigger conditions, and success data from real projects.
+
+## Seeding Script
+
+The Command nodes and trigger relationships need to be seeded into Neo4j. Create a script that:
+
+1. Reads all commands/*.md files
+2. Extracts: name, description, allowed-tools, category
+3. Generates JTBD statements from the command's purpose
+4. Creates Command nodes with relationships to:
+   - Frameworks they follow (from methodology chaining rules)
+   - VentureStages they're relevant at (from problem-types.md)
+   - SignalTypes that trigger them (from proactive-intelligence patterns)
+5. Runs the Cypher via `mcp__neo4j-brain__write_neo4j_cypher`
+
+Script location: `scripts/seed-brain-commands.cjs`
+
+## The Intelligence Hierarchy (Updated)
+
+```
+LEVEL 3 (Brain + Room + Signals):
+  brain_proactive_command (Pattern 10d)
+  Multi-hop: frameworks -> commands -> triggers -> JTBD
+  Calibrated from 100+ real projects
+  Knows which sequences ACTUALLY produced results
+
+LEVEL 2 (Room + Local Fabric):
+  Larry's JTBD provoked suggestions (every 3-7 turns)
+  KuzuDB Tensions, Bottlenecks, Surprises, Convergences
+  Good but not cross-venture calibrated
+
+LEVEL 1 (No Room):
+  Generic stage-based defaults
+  methodology/index.md routing table
+  Better than nothing
+```
+
+Brain-connected users get LEVEL 3: the Brain literally tells Larry which command to suggest, when, and why -- backed by data from 100+ real ventures. The JTBD framing is baked into the Command node itself. Larry just reads it and presents it naturally.