By name: pod-reasoning description: > Deep architectural reasoning using the o4-mini frontier model. Evaluates trade-offs, recommends patterns, and produces decision rationales for scalability, security, and maintainability. Use when: "design system", "compare approaches", "architecture decision", "complex problem", "trade-offs". license: MIT compatibility: "Requires: Azure OpenAI O4-Mini deployment" metadata: author: peterpodj category: architecture version: "1.0" keywords: - reasoning - architecture - design - trade-offs - scalability - security allowed-tools:
- LLM(o4-mini)
- Read
Reasoning Skill
When to Activate
- System design, architecture choices, trade-off analysis
- Scalability/security/latency considerations
- Migration strategy, pattern selection, risk assessment
Core Patterns (Level 2)
- Architecture template: templates/architecture.md
- Decision record: templates/adr.md
- Risk matrix: templates/risk_matrix.md
Guidance
- Produce clear options with pros/cons and explicit risks.
- Provide a recommended option with rationale and next steps.
- Align with PETERPOD standards (async-first, tests, tracing, Pydantic config).
Step-by-Step Workflow
- Clarify constraints (SLOs, data volume, latency, budget).
- Propose 2-3 options with trade-offs.
- Recommend one option; list risks and mitigations.
- Outline implementation phases and test strategy.
Full Reference (Level 3)
See references/REFERENCE.md for detailed patterns and examples.