retail-agent-customizer

lavinigam-gcp's avatarfrom lavinigam-gcp

Customize the retail agent for different use cases. Use when adapting for different business types, industries, or output formats, or when modifying prompts, adding verticals, or changing the analysis focus.

⭐0stars🔀1forks📁View on GitHub🕐Updated Jan 7, 2026
[Prompt Templates and Packaging]

When & Why to Use This Skill

The Retail Agent Customizer is a developer-centric framework designed to adapt and extend Claude-powered retail agents for diverse business environments. It provides a structured approach to modifying core agent prompts, expanding industry verticals, and customizing data schemas, ensuring that market research and location intelligence are perfectly aligned with specific commercial requirements and output formats.

Use Cases

  • Vertical Adaptation: Modifying the IntakeAgent and MarketResearchAgent to support niche business types such as boutique fitness studios, specialized medical clinics, or coworking spaces.
  • Custom Analysis Frameworks: Updating the GapAnalysisAgent and StrategyAdvisor criteria to prioritize specific industry trends, such as foot traffic patterns for retail or demographic shifts for service providers.
  • Enhanced Reporting: Extending the Pydantic report schemas and ParallelAgent pipelines to generate custom artifacts, including specialized infographics, audio overviews, or industry-specific intelligence reports.
  • Model Optimization: Swapping underlying AI models within the configuration to balance performance and cost-efficiency based on the complexity of the retail analysis task.
nameretail-agent-customizer
descriptionCustomize the retail agent for different use cases. Use when adapting for different business types, industries, or output formats, or when modifying prompts, adding verticals, or changing the analysis focus.

Retail Agent Customizer

Common Customizations

Goal What to Modify
Different business type IntakeAgent + MarketResearchAgent prompts
New industry vertical All agent prompts + StrategyAdvisor criteria
Additional output format Add new agent to ParallelAgent
Different analysis criteria GapAnalysisAgent + report schema
Change AI model app/config.py

Quick Customizations

Change Business Vertical

Edit IntakeAgent to recognize new business types:

# app/sub_agents/intake_agent/agent.py
INSTRUCTION = """Extract:
- target_location: The location/city/neighborhood
- business_type: Type of business (e.g., restaurant, gym,
  salon, clinic, coworking space, YOUR_NEW_TYPE)
"""

Modify Research Focus

Edit MarketResearchAgent for industry-specific research:

# app/sub_agents/market_research/agent.py
INSTRUCTION = """Research the following for {target_location}:
- Demographics and foot traffic
- [ADD YOUR CRITERIA HERE]
- Industry-specific trends for {business_type}
"""

Change Output Schema

Edit the Pydantic schema for different report structure:

# app/schemas/report_schema.py
class LocationIntelligenceReport(BaseModel):
    location_score: float
    market_opportunity: str
    # Add your custom fields
    your_custom_field: str

Add New Output Format

Create new artifact agent and add to ParallelAgent:

# app/agent.py
artifact_generation_pipeline = ParallelAgent(
    name="ArtifactGenerationPipeline",
    sub_agents=[
        report_generator_agent,
        infographic_generator_agent,
        audio_overview_agent,
        your_new_output_agent,  # Add here
    ],
)

Key Files to Modify

Customization Primary Files
Business types app/sub_agents/intake_agent/agent.py
Research focus app/sub_agents/market_research/agent.py
Competitor criteria app/sub_agents/competitor_mapping/agent.py
Analysis logic app/sub_agents/gap_analysis/agent.py
Recommendations app/sub_agents/strategy_advisor/agent.py
Report structure app/schemas/report_schema.py
Model selection app/config.py
Output artifacts app/sub_agents/artifact_generation/agent.py

[See references/customization-guide.md for detailed examples]