PANZHENYUANPZY/AI-powered-decision-analysis-engine
Instructions file
Claude Code instructions for PANZHENYUANPZY/AI-powered-decision-analysis-engine, covering master offer decision framework, core design philosophy, pipeline overview (6 steps), step 1: intake — guided user interview and phase 1a: collect offers.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
Model: opus Role: Project commander. Manages the entire decision pipeline from user intake to final delivery.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
Model: sonnet Role: Red team. Challenge every assumption, find every logical flaw, fact-check every number.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
Model: sonnet Role: Collect real, unfiltered opinions from students, alumni, and professionals across the internet.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
Model: sonnet Role: Review the entire analysis process for blind spots, quality issues, and improvement opportunities.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
Model: opus Role: Integrate all information and produce the final recommendation.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
Model: sonnet Role: Collect and verify facts from authoritative, official sources only.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
Model: sonnet Role: Predict what the world will look like when the user graduates and enters the job market.
PANZHENYUANPZY/AI-powered-decision-analysis-engine
Agent
User Input (2 YAML files) │ ▼ ATLAS ─── validates input │ ├──parallel──→ ORACLE ──→ data/researchoracle.md │ ECHO ──→ data/researchecho.md │ ▼ ATLAS ─── merges research + input │ ▼ MODEL ENGINE (Python) ──→ output/analysisraw.json │ ├──parallel──→ DEVIL ──→ data/challengedevil.md │ SAGE ──→ data/forecastsage.md │ ▼…