Systematic Find-Prove-Evidence-Fix framework for complex system analysis and intervention. Use when debugging failures, investigating incidents, analyzing performance issues, or optimizing complex systems.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Um comando direto ignora o prompt de revisão. Verifique a origem antes de executá-lo.
Systematic Find-Prove-Evidence-Fix framework for complex system analysis and intervention. Use when debugging failures, investigating incidents, analyzing performance issues, or optimizing complex systems.
FPEF Evidence Analyzer
Overview
Rigorous evidence-based analysis framework that systematically finds problems, proves root causes, gathers supporting evidence, and implements targeted fixes for complex technical and business systems.
Prerequisites
System or problem description
Access to relevant data sources (logs, metrics, code)
Basic understanding of the domain being analyzed
What This Skill Does
Find: Systematic identification of anomalies and potential issues
Prove: Causal analysis to establish root cause relationships
Evidence: Comprehensive evidence collection and validation
Fix: Targeted interventions with verification of effectiveness
{"problem_description":"Clear description of the issue","scope":"Systems, components, or processes affected","timeline":"When the problem started and duration","severity":"critical|high|medium|low","business_impact":"Revenue, users, operations affected","data_sources":["logs","metrics","traces","code"],"constraints":["time","budget","access","expertise"]}
resources/evidence/code-analysis.template - Code review and analysis
Success Metrics
Analysis Quality Metrics
Root Cause Identification Accuracy: % of fixes that resolve the actual problem
Time to Resolution: Average time from problem detection to fix implementation
Evidence Completeness: % of required evidence successfully collected
Fix Effectiveness: % reduction in problem occurrence after intervention
Process Metrics
Hypothesis Validation Rate: % of hypotheses confirmed or refuted
Evidence Reliability Score: Average confidence level in collected evidence
Cross-functional Collaboration: Number of departments successfully engaged
Knowledge Transfer: % of insights documented and shared
Troubleshooting
Issue: Insufficient Evidence
Symptoms: Cannot reach 95% confidence in root cause
Solution:
Review resources/fpef-config.json for additional data sources
Run ./scripts/expand-scope.sh to broaden investigation scope
Use expert interviews for qualitative evidence
Implement controlled experiments for causal proof
Issue: Multiple Competing Hypotheses
Symptoms: Several equally likely root causes identified
Solution:
Run ./scripts/hypothesis-prioritization.sh based on impact and likelihood
Implement parallel investigation tracks
Use controlled experiments to test each hypothesis
Apply Occam's razor principle for simplicity preference
Issue: Fix Implementation Resistance
Symptoms: Teams reluctant to implement proposed fixes
Solution:
Generate resources/stakeholder-analysis.md for change management
Create detailed implementation timelines with milestones
Provide clear ROI calculations for proposed changes
Setup pilot programs to demonstrate fix effectiveness
Integration with Other Skills
Complementary Skills
UltraPlan: Use findings to improve future planning processes
MCP Manager: Integrate with monitoring systems for continuous evidence collection
Multi-Agent Systems: Coordinate analysis across technical teams
External System Integration
# Connect to monitoring and observability platforms
./scripts/integrate-monitoring.sh --platform datadog,prometheus,grafana
# Connect to incident management systems
./scripts/integrate-incident.sh --system pagerduty,opsgenie
# Connect to development and deployment systems
./scripts/integrate-devops.sh --tools jenkins,gitlab,circleci