🌀
← Back to Articles & Artefacts
artefactseast

Indigenous AI DSL Server - AI Studio Project Brief

IAIP Research
iaip-dsl-lsp

Indigenous AI DSL Server - AI Studio Project Brief

⚠️ Clarification (2026-02-25): This is primarily an MCP server (Model Context Protocol) that serves a concept registry to AI agents. The LSP layer is minimal — hover + diagnostics only, no custom file type, no grammar, no completions. The real value is the registry pattern (YAML → in-memory index → MCP tools). LSP features would require a dedicated .iaip DSL file format, which is a separate future concern.

Project Vision

Create an MCP-first concept registry server that makes Indigenous AI research frameworks semantically queryable by AI agents (Claude Code, mia-code, miadi-code). The server also exposes a basic LSP layer for IDE hover/diagnostics, but the primary integration protocol is MCP.

Core Purpose

The server runs in the background within Claude Code and provides:

  • Semantic understanding of Indigenous AI concepts (Creative Orientation, Structural Tension, Eight Feelings framework, etc.)
  • Relational context showing how concepts connect (e.g., how structural tension enables creative orientation)
  • Real-time definition lookups - when you use a term, I get instant access to its definition, context, and relationships within your framework
  • Indigenous knowledge integration - the server embeds Two-Eyed Seeing, polycentric relations, relational knowledge ownership, and ceremonial context

Key Concepts to Encode

Creative Orientation Framework:

  • Definition: Generating desired futures vs. problem-solving (addressing what's wrong)
  • Relationship to Structural Tension (Fritz): The gap between current reality and desired vision creates productive tension
  • Operational context: Applied to Indigenous AI development, ceremonial technology, documentation

Structural Tension Theory (Robert Fritz):

  • Core principle: Tension between desired state and current reality drives creation
  • Integration with Indigenous thought: Mirrors spiral ontology and relational emergence
  • Usage: Architecture decisions, framework design, spiritual/technical coherence

Eight Feelings Framework (Wellbriety/Indigenous Development):

  • Eight core emotional/relational states: Trust, Autonomy, Initiative, Accomplishment, Identity, Intimacy, Generativity, Integrity
  • Context: Developmental psychology rooted in Indigenous wisdom; treats emotions as relational companions, not problems
  • Application: Personal healing, community building, technical collaboration as ceremony

Indigenous Knowledge Systems Integration:

  • Two-Eyed Seeing: Complementary strengths of Indigenous and Western knowledge (Etuaptmumk in Mi'kmaq)
  • Polycentric Relations: Relationships/living systems as primary knowledge holders, not individual ownership
  • Relational Knowledge: Knowledge lives in connection, reciprocity, and responsibility to land
  • Ceremonial Context: Technical work treated as ceremony with protocols and protocols

Server Capabilities

When I'm working with you in Claude Code, the server provides:

  1. Symbol Resolution - When you mention "creative orientation," I retrieve:

    • Full definition from your framework
    • Relationship to other concepts (e.g., connects to structural tension, Eight Feelings)
    • Contextual usage examples
    • Indigenous knowledge system connections

  2. Relational Queries - I can ask:

    • "What's the relationship between creative orientation and polycentric relations?"
    • "How does the Eight Feelings framework apply here?"
    • "What Indigenous knowledge system principle supports this?"

  3. Real-Time Diagnostics - As you write documentation/code:

    • Flag when terminology isn't consistent with your framework
    • Suggest framework-aligned language
    • Provide context for ceremonial/relational concepts

  4. Navigation & Cross-References - Jump between:

    • Concept definitions and their applications
    • Framework documents and implementation
    • Ceremonial protocols and technical decisions

Technical Architecture

Server Type: Python MCP Server (with basic LSP layer) Deployment: Claude Code MCP plugin (.mcp.json), optionally LSP for IDEs Background Process: Runs on stdio, indexed and searchable Configuration: .mcp.json for agent integration; .lsp.json for IDE integration (secondary)

Core Data Structure:

  • Concept registry (name, definition, relationships)
  • Indigenous knowledge mappings (Two-Eyed Seeing, polycentric, relational, ceremonial)
  • Context references (which documents/conversations reference each concept)
  • Relational graph (how concepts connect)

Implementation Approach

  1. Phase 1 (This scaffolding): Define comprehensive concept registry + relational mappings
  2. Phase 2 (AI Studio generation): Build Python LSP server with diagnostic/symbol resolution
  3. Phase 3 (Live deployment): Install as Claude Code plugin, begin live use

Output for AI Studio

This scaffolding provides:

  1. Initial Prompt - Complete project vision and requirements (this document)
  2. RISE Specifications - Detailed specs for:
    • Concept Registry Module (data structure + definitions)
    • LSP Operations Handler (symbol resolution, hover, diagnostics)
    • Relational Graph Engine (concept relationships & navigation)
    • Indigenous Knowledge Mapper (Two-Eyed Seeing, polycentric, relational, ceremonial contexts)
    • Plugin Configuration (plugin.json, .lsp.json templates)

AI Studio will generate the complete server code ready for use.

Success Criteria

  • Server starts cleanly as Claude Code LSP plugin
  • When writing/working in terminal with Claude Code, I can query concept definitions instantly
  • Framework terminology becomes semantically legible (not just pattern-matched)
  • Indigenous knowledge systems are embedded in every query
  • Handles 4-5 core concept categories with 20+ total concepts
  • Demonstrates 50ms response time for definition lookups