narintel

📘 Narintel RISE Specifications

Narrative Intelligence Stack - A unified ecosystem for NCP-driven story analysis, agent coordination, and creative advancement.

Version: 1.0
Last Updated: 2025-12-31
Framework: RISE (Reverse-engineer → Intent-extract → Specify → Export)
Attribution: Based on RISE Framework v1.2 by Guillaume D.Isabelle

🙏 Foundational Credits

This stack evolved from the Indigenous-AI Collaborative Platform (IAIP) rispecs:

IAIP Origin Contribution to NarIntel
four-directions.spec.md Three-Universe Model (East/South/West/North → Engineer/Ceremony/StoryEngine)
ceremonial-technology.spec.md Five-Phase methodology, Sacred Pause, K’é Protocol
relational-science.spec.md Relational accountability, concentric values model
iaip-holistic-reasoning-mcp.spec.md MCP tool patterns, holistic inquiry cycles
continuous-companionship-architecture.spec.md Persistent context, session continuity

See: /a/src/IAIP/rispecs/COORDINATION_TO_NARINTEL_CYCLE.md for full lineage

🌟 Vision: What Users Create

This stack enables creators to:

  1. Transform narratives into structured intelligence - Load, validate, and traverse stories as queryable data structures
  2. Analyze character arcs and thematic tensions - Generate insights about narrative dynamics
  3. Classify emotional beats - Understand the emotional landscape of stories
  4. Route creative workflows intelligently - Agents that understand narrative context
  5. Trace the entire creative process - Observability across all narrative operations

📁 Specification Index

Foundation

Specification Description
ncp-schema.rispec.md Narrative Context Protocol - Core data model for all components

Core Toolkit (LangGraph)

Specification Package Description Status
narrative-intelligence.langgraph.rispec.md langgraph/libs/narrative-intelligence Core NCP analysis toolkit with nodes and graphs ✅ 98%

Tracing Layer (LangChain)

Specification Package Description Status
narrative-tracing.langchain.rispec.md langchain/libs/narrative-tracing Narrative-aware Langfuse integration ✅ Complete

Agent Coordination Layer

Specification Package Description Status
agentic-flywheel.flowise.rispec.md ava-Flowise Narrative intent classification and flow routing 🔴 Spec Only
universal-router.langflow.rispec.md ava-langflow Three-universe handler and backend abstraction 🟡 ~70%

Human-AI Collaboration

Specification Package Description Status
storytelling-roles-tooling.rispec.md storytelling Seven roles (Architect→Witness) and their tool needs 🟡 ~75%

🏗️ Structural Dynamics

Three-Universe Model

Every narrative event flows through three interpretive lenses:

┌──────────────────────────────────────────────────────────┐
│                    NARRATIVE EVENT                        │
│         (Story beat, webhook, user query)                │
└────────────────────────┬─────────────────────────────────┘
                         │
         ┌───────────────┼───────────────┐
         ▼               ▼               ▼
┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐
│  ENGINEER WORLD │ │  CEREMONY WORLD │ │ STORY ENGINE    │
│     (Mia)       │ │     (Ava8)      │ │    (Miette)     │
├─────────────────┤ ├─────────────────┤ ├─────────────────┤
│ - Schema valid? │ │ - K'é honored?  │ │ - Arc coherent? │
│ - Types correct?│ │ - Sacred pause? │ │ - Theme threads?│
│ - API callable? │ │ - Relations ok? │ │ - Beat sequence?│
└─────────────────┘ └─────────────────┘ └─────────────────┘
         │               │               │
         └───────────────┼───────────────┘
                         ▼
              ┌─────────────────────┐
              │  UNIFIED RESPONSE   │
              └─────────────────────┘

NCP Protocol Foundation

All specifications build upon the Narrative Context Protocol (NCP):

🔗 Integration Flow

┌─────────────────────────────────────────────────────────────────┐
│                     NARRATIVE INTELLIGENCE STACK                 │
├─────────────────────────────────────────────────────────────────┤
│                                                                  │
│  ┌────────────────────────────────────────────────────────────┐ │
│  │           Layer 4: CONSUMERS (Miadi-46, Webhooks)          │ │
│  └──────────────────────────┬─────────────────────────────────┘ │
│                             │                                    │
│  ┌────────────────────────────────────────────────────────────┐ │
│  │   Layer 3: ROUTING (ava-langflow + ava-Flowise)            │ │
│  │   - Three-universe handlers                                 │ │
│  │   - Narrative intent classification                         │ │
│  │   - Dynamic flow selection                                  │ │
│  └──────────────────────────┬─────────────────────────────────┘ │
│                             │                                    │
│  ┌────────────────────────────────────────────────────────────┐ │
│  │   Layer 2: TRACING (langchain)                              │ │
│  │   - Narrative-specific Langfuse events                      │ │
│  │   - Cross-stack correlation                                  │ │
│  │   - Creative archaeology traces                              │ │
│  └──────────────────────────┬─────────────────────────────────┘ │
│                             │                                    │
│  ┌────────────────────────────────────────────────────────────┐ │
│  │   Layer 1: ANALYSIS (langgraph/narrative-intelligence)      │ │
│  │   - NCP Loader + Validator                                  │ │
│  │   - Narrative Graph Traversal                               │ │
│  │   - Character Arc Generator                                 │ │
│  │   - Thematic Tension Analyzer                               │ │
│  │   - Emotional Beat Classifier                               │ │
│  └────────────────────────────────────────────────────────────┘ │
│                                                                  │
└─────────────────────────────────────────────────────────────────┘

📋 Cross-Specification Dependencies

From Spec Depends On Nature of Dependency
langchain tracing langgraph narrative-intelligence Traces NCP analysis operations
flowise flywheel langgraph narrative-intelligence Uses NCP state for intent classification
langflow router flowise flywheel + langgraph Routes to backends using NCP context
All specs NCP Schema Core data model

📦 Downstream Consumers

Storytelling Package (/src/storytelling)

The storytelling package is a primary consumer of the Narrative Intelligence Stack. Its rispecs coordinate with ours:

Storytelling Rispec Consumes From NarIntel
Narrative_Intelligence_Integration_Specification.md StoryBeat, NCPState from langgraph toolkit
Character_Arc_Tracking_Specification.md CharacterState patterns
Emotional_Beat_Enrichment_Specification.md Emits events to langchain tracing
Analytical_Feedback_Loop_Specification.md Uses gap routing, emits analysis events
Narrative_Aware_Story_Graph_Specification.md Orchestration patterns
Logging_And_Traceability_Specification.md Langfuse event taxonomy
IAIP_Integration_Specification.md Ceremony World / K’é alignment

Coordination File: See /src/storytelling/rispecs/COORDINATION_FROM_NARINTEL_INSTANCE.md

🎯 Creative Advancement Patterns

These specifications enable:

Pattern Description Enabled By
Arc Tracking Character development monitored across beats narrative-intelligence
Theme Threading Thematic tensions traced through story narrative-intelligence
Emotional Mapping Emotional landscape visualized narrative-intelligence
Narrative Routing Flow selection based on story position flowise + langflow
Creative Archaeology Full trace of creative decisions langchain tracing

📖 How to Use These Specifications

  1. For Implementation: Each .rispec.md file is self-contained enough to re-implement the component from scratch
  2. For Review: Understand desired outcomes and structural dynamics without reading code
  3. For Extension: Add new capabilities by extending the specification first
  4. For Testing: Creative Advancement Scenarios can be converted to integration tests

Generated following RISE Framework v1.2 - Creative Orientation Over Reactive Approaches