@cloudkinetix/bmad-enhanced/expansion-packs/ck-llm-agent-dev/tasks/create-agent-spec.md

Cloud-Kinetix enhanced fork of BMAD-METHOD - Breakthrough Method of Agile AI-driven Development with robust versioning and unified validation.

# Research-Driven Agent Specification Creation

This task guides the creation of comprehensive AI agent specifications through research-driven methodology, focusing on discovering current specification best practices rather than prescriptive static templates.

## Research-First Specification Development

[[LLM: Begin by researching current AI agent specification methodologies, frameworks, and industry standards. Understand the specific specification requirements and documentation landscape before creating agent specifications.]]

### 1. Research Specification Approaches

**Specification Framework Research Areas**:

- Current AI agent specification methodologies and documentation standards
- Latest developments in AI system requirements engineering
- Industry-standard specification frameworks and templates
- Best practices for AI agent requirements gathering and analysis
- Stakeholder engagement patterns for AI system specification

**Requirements Engineering Research**:

- Requirements elicitation techniques for AI systems
- Functional and non-functional requirement patterns for AI agents
- Use case modeling approaches for conversational AI systems
- Behavioral specification methodologies for AI agents
- Safety and compliance requirement frameworks for AI applications

### 2. Research-Based Implementation Strategy

[[LLM: Based on your research findings, create agent specifications using current best practices. Focus on:

1. **Specification Framework Selection**: Choose specification frameworks based on researched industry standards and project requirements
2. **Requirements Methodology**: Apply requirements engineering using current elicitation and analysis techniques
3. **Documentation Structure**: Structure specifications using research-informed documentation patterns
4. **Validation Approach**: Validate specifications using current verification and validation methodologies
5. **Stakeholder Engagement**: Engage stakeholders using research-backed requirements gathering approaches

Document your specification creation choices and rationale based on the research conducted.]]

### 3. Specification Development Framework

**Research Current Specification Approaches**:

- Investigate agent capability definition methodologies
- Study behavioral specification techniques for AI systems
- Research technical requirement documentation patterns for AI agents
- Analyze safety and compliance specification approaches for AI applications

**Implementation Areas**:

- Establish specification structure using researched documentation frameworks
- Define agent capabilities based on current requirement analysis methodologies
- Document behavioral requirements using research-informed specification patterns
- Specify technical requirements using current AI system documentation best practices

### 4. Validation and Review Process

**Research Validation Methodologies**:

- Investigate specification validation techniques for AI systems
- Study stakeholder review processes for AI agent specifications
- Research specification iteration and refinement methodologies
- Analyze specification maintenance approaches for evolving AI systems

**Implementation Validation**:

- Apply research-backed validation methodologies to ensure specification completeness
- Use current review techniques to validate specification accuracy and feasibility
- Implement stakeholder feedback processes based on researched best practices
- Establish specification maintenance processes using current documentation patterns

---

**Note**: This task emphasizes research-driven agent specification creation over prescriptive static templates. Always research current specification standards and adapt to your specific agent development requirements and organizational context.