agentic-data-stack-community
Version:
AI Agentic Data Stack Framework - Community Edition. Open source data engineering framework with 4 core agents, essential templates, and 3-dimensional quality validation.
182 lines (161 loc) • 6.31 kB
YAML
workflow:
id: community-analytics
name: Community Analytics Workflow
description: >-
Streamlined analytics workflow for community users with basic agent collaboration
and standard validation. Supports exploratory and business analysis projects.
type: greenfield
project_types:
- exploratory-analysis
- business-intelligence
- performance-analytics
- customer-insights
sequence:
- step: requirements_gathering
agent: data-product-manager
action: gather-requirements
creates: analysis-requirements.md
duration: 0.5 days
notes: |
Gather business requirements and define analysis objectives:
- Business context and objectives
- Key stakeholders and success criteria
- Data sources and availability assessment
- Timeline and resource constraints
SAVE OUTPUT: Copy final analysis-requirements.md to your project's docs/ folder.
- step: data_profiling
agent: data-analyst
action: profile-data
creates: data-profile-report.md
requires: analysis-requirements.md
duration: 0.5 days
notes: |
Conduct data profiling and discovery:
- Data quality assessment
- Schema understanding and relationship mapping
- Missing data patterns and anomaly detection
- Statistical summaries and distribution analysis
- Data readiness evaluation for analysis goals
- step: analysis_execution
agent: data-analyst
action: analyze-data
creates: analysis-results.md
requires: [analysis-requirements.md, data-profile-report.md]
duration: 1-2 days
notes: |
Execute data analysis:
- Statistical analysis and hypothesis testing
- Pattern discovery and trend analysis
- Business insights and recommendations
- Data visualization and charts
- step: dashboard_creation
agent: data-analyst
action: create-dashboard
creates: analytical-dashboard
requires: analysis-results.md
duration: 1 day
notes: |
Create dashboard and visualizations:
- Interactive charts and visualizations
- Key metrics and KPI displays
- User-friendly navigation and filters
- Export and sharing capabilities
- step: quality_validation
agent: data-quality-engineer
action: validate-data-quality
validates: [data-profile-report.md, analysis-results.md, analytical-dashboard]
duration: 0.5 days
notes: |
Quality validation of analysis outputs:
- Data accuracy validation in analytical results
- Calculation verification and testing
- Dashboard functionality testing
- Cross-validation with known business metrics
- step: stakeholder_review
agent: data-product-manager
action: define-metrics
requires: [analysis-results.md, analytical-dashboard]
duration: 0.5 days
notes: |
Stakeholder validation and sign-off:
- Present findings to business stakeholders
- Validate business impact and actionability
- Document success metrics and KPIs
- Plan implementation and next steps
validation_gates:
- gate: requirements_validation
criteria:
- Business objectives clearly defined and measurable
- Data sources identified and accessible
- Success criteria established with stakeholders
- Resource requirements validated and approved
- gate: analysis_validation
criteria:
- Statistical methods appropriate for data and objectives
- Quality validation passes framework standards
- Insights are actionable and business-relevant
- Visualizations follow best practices
- gate: delivery_validation
criteria:
- All deliverables pass quality checks
- Stakeholder approval documented
- Success metrics established
- Implementation plan documented
success_criteria:
business:
- Key stakeholders can articulate value and next steps
- Insights directly address business objectives
- Recommendations are actionable and prioritized
- Success metrics are established and trackable
technical:
- Dashboard performance meets requirements
- Data accuracy meets business quality standards
- Analysis is reproducible and documented
- User adoption targets are achievable
analytical:
- Statistical analysis is methodologically sound
- Business insights are supported by evidence
- Visualizations effectively communicate findings
- Documentation enables reproducibility
workflow_variants:
exploratory_analysis:
focus: Open-ended data exploration and hypothesis generation
duration: 2-3 days
additional_emphasis:
- Pattern discovery and trend identification
- Hypothesis generation and validation
- Exploratory data visualization
business_intelligence:
focus: KPI monitoring and business performance analysis
duration: 3-4 days
additional_emphasis:
- KPI definition and calculation
- Performance dashboards
- Business reporting automation
customer_insights:
focus: Customer behavior analysis and segmentation
duration: 3-5 days
additional_emphasis:
- Customer segmentation analysis
- Behavior pattern identification
- Personalization opportunities
escalation_procedures:
- condition: Data quality issues preventing reliable analysis
action: Escalate to Data Quality Engineer and Data Engineer
timeline: Within 2 hours
- condition: Technical analysis beyond analyst capability
action: Consider external expertise or simplified approach
timeline: Within 4 hours
- condition: Stakeholder requirement conflicts or scope changes
action: Escalate to Data Product Manager and Business Sponsors
timeline: Within 1 business day
post_completion_activities:
- activity: Dashboard usage monitoring
frequency: Weekly for first month, then monthly
responsible: Data Analyst
- activity: Business impact tracking
frequency: Monthly for first quarter
responsible: Data Product Manager
- activity: Analysis quality review
frequency: Quarterly
responsible: Data Quality Engineer