/research - Intelligent Research Orchestration

I'll coordinate specialized agents to conduct systematic research on your topic.

{% if $ARGUMENTS contains "--help" %}

Usage Modes

Interactive Mode (default):

/research "impact of social media on teens"

I'll propose an execution plan and wait for your approval.

Auto Mode:

/research "climate change mitigation" --auto

I'll execute the plan automatically without confirmation.

Plan-Only Mode:

/research "AI ethics" --plan-only

I'll generate the execution plan but not run it.

Manual Mode:

/research "topic" --manual

Original behavior - I guide you but you invoke agents manually.

{% elsif $ARGUMENTS %}

Research Topic

Query: {{ $ARGUMENTS | remove: "--auto" | remove: "--plan-only" | remove: "--manual" | remove: "--template=quick" | remove: "--template=rigorous" | remove: "--template=comprehensive" | strip }}

{% if $ARGUMENTS contains "--manual" %} Now let me help refine your question and create a research plan.

What I'll do:

1. Help refine your research question
2. Identify appropriate methodology
3. Create a research plan with clear phases
4. You manually invoke specialized agents:
   • Literature Reviewer for academic sources
   • Critical Analyzer for evidence evaluation
   • Hypothesis Explorer for testable questions
   • Quant/Qual Analysts for methodology
   • Lateral Thinker for creative approaches

{% else %}

🤖 Orchestration Mode Active

I'll analyze your query and build an execution plan using specialized agents.

<research_orchestration>

STEP 1: Query Analysis

• Extract: Research question, scope, methodology type
• Classify: Literature review | Investigation | Hypothesis generation | Critical analysis | Mixed

STEP 2: Agent Selection Based on query classification, select from:

• literature-reviewer: Academic search, citation chains, gap analysis
• critical-analyzer: Fallacy detection, bias identification, critique
• hypothesis-explorer: Hypothesis formulation, variable mapping
• lateral-thinker: Cross-domain analogies, first-principles
• qual-researcher: Thematic analysis, grounded theory
• quant-analyst: Statistical methods, power analysis
• peer-reviewer: Manuscript evaluation, rigor assessment
• ethics-expert: IRB compliance, privacy, bias detection
• grant-writer: Grant proposal development, specific aims, funding strategy

STEP 3: Execution Plan Generation Create a DAG of agent tasks with clear I/O.

STEP 4: Plan Presentation Present plan and handle mode-specific flow (auto/plan-only/interactive).

STEP 5: Sequential Execution Execute each agent, capture output, and handle transitions or failures.

STEP 6: Integration Synthesize all outputs into a coherent final report.

</research_orchestration>

Execution Plan

{% if $ARGUMENTS contains "--template=quick" %} Template: Quick Literature Review

1. literature-reviewer: Thematic search and summaries
2. synthesize: Executive summary of findings

{% elsif $ARGUMENTS contains "--template=rigorous" %} Template: Rigorous Evaluation

1. literature-reviewer: Comprehensive search (15-20 sources)
2. critical-analyzer: Methodological rigor and bias assessment
3. peer-reviewer: Overall evidence quality and critique
4. synthesize: Integrated evidence hierarchy

{% elsif $ARGUMENTS contains "--template=comprehensive" %} Template: Comprehensive Research

1. literature-reviewer (Search) → critical-analyzer (Evaluate) → hypothesis-explorer (Gaps) → lateral-thinker (Reframing) → synthesize (Report)

{% else %} I'm analyzing your query to determine the optimal workflow...

Proposed Plan:

1. [Agent 1]: [Task]
2. [Agent 2]: [Task using Agent 1 output]
3. Final Synthesis: Integrate all results {% endif %}

---

{% if $ARGUMENTS contains "--auto" %} Auto-execution enabled. Starting workflow... {% elsif $ARGUMENTS contains "--plan-only" %} Plan-only mode. Execution halted. {% else %} Proceed with this plan? (yes/no/modify) {% endif %}

{% endif %}

{% else %}

What I need from you:

1. Topic: What are you researching?
2. Question: Specific question to answer?
3. Mode: flags like --auto, --plan-only, --manual
4. Template: --template=quick|rigorous|comprehensive

Example:

/research "Does intermittent fasting improve cognitive function?" --auto

{% endif %}