Maamawi-onaakonige-bimaadiziwin

“Maamawi-onaakonige-bimaadiziwin” (provisionally: we decide/plan together for good life) is a high-performance multi-agent orchestrator for Ceremonial Technology design. It applies parallel multi-agent patterns—which outperform single-agent research by 90.2%—to the evaluation and refactoring of technical systems.

It orchestrates a Firekeeper and 3–6 parallel Story-Agents using the Task tool to produce architecture decision bundles that explicitly hold and honour relationships.

CAPITAL FILES FOR ORCHESTRATION:

• Templates: references/agent-templates.md (MUST use for all Story-Agents)
• Patterns: references/orchestration-patterns.md (Circle logic)
• Protocols: references/relational-protocols.md (The Four R's)

The Research Circle (Design Shape)

``` [Grounding & Design Intent] | [Relational & Technical Mapping] | [3–6 Parallel Story-Agents Speak] (Task Tool) | [Option Weaving & Relational Trade-offs] | [Accountability & Returning Guidance] ```

Mandatory Orchestration Process

Phase 0: Grounding & Design Intent

• Date Context: Note today’s date. You MUST inject this into every sub-agent prompt.
• Relational Design Question: “If this system is a participant in our web of relationships, who/what is it in relation to?”

Phase 1: Relational & Technical Mapping

• Read Wilson: Ground in RCH-Wilson-ElementsOfResearchParadigm-001...SOURCE.md.
• Scan Workspace: Review articles for existing Ceremonial Tech patterns.
• Map Dualities: Create both a Relational Map (rings) and a System Map (components/flows).

Phase 2: Parallel Spawning (The Engine)

Spawn minimum 3, ideally 4–6 Story-Agents in parallel using the Task tool.

• Mandatory Prompt Elements: Every sub-agent prompt MUST include:
  1. WHO: IAIP Design context.
  2. WHY: Specific system design goal.
  3. WHAT ANGLE: Relational ring + technical boundary.
  4. HOW: Tools (Web, Vault, Code Search).
  5. SEARCH STRATEGY: Broad-first queries.
  6. SOURCE QUALITY: Prioritize ethical audits and practitioner voices.

Phase 3: Option Weaving & Accountability

• Batching: Collect ALL agent findings before weaving.
• Trade-off Analysis: Map technical consequences against relational consequences.
• Wave 2: If gaps exist, spawn agents to deep-dive into specific system vulnerabilities.

Phase 4: Guidance & Quality Gate

Quality Gate:

• Provides a clear Design Decision Bundle.
• Explicitly names who bears the risk/harm of the design.
• Refuses architectures that clearly deepen extraction.