How prajwal471/captain_cool works
Captain Cool is an interactive IPL match-strategy assistant. A user enters match conditions in the CLI; the orchestrator gathers a stats report, gets an initial tactical proposal, requests a Devil’s Advocate challenge, then asks the Strategist for a final decision and displays the debate. The Stats Analyst can use mock live-context and historical-stat tools. The README describes these capabilities; the sampled source confirms the orchestration and mock tools, not any connection to real match-data services.
Main entry points: orchestrator.py Pipeline orchestrator · main.py Match CLI · agents.py Devil's Advocate · tools.py Live data tool
Components 10
User interface
- Match CLI
main.py
Strategy debate
- Pipeline orchestrator
orchestrator.py - Stats Analyst
agents.py - Initial proposal
orchestrator.py - Devil's Advocate
agents.py - Final decision
orchestrator.py
Match data tools
- Live data tool
tools.py - Historical stats tool
tools.py
Other components
- Match user
- Gemini service
Connections 14
- Match user → Match CLI: enters match state
- Match CLI → Pipeline orchestrator: passes match state
- Pipeline orchestrator → Gemini service: requests analysis
- Pipeline orchestrator → Live data tool: registers tool
- Pipeline orchestrator → Historical stats tool: registers tool
- Gemini service → Live data tool: dispatches tool call
- Gemini service → Historical stats tool: dispatches tool call
- Pipeline orchestrator → Initial proposal: requests proposal
- Initial proposal → Gemini service: generates proposal
- Pipeline orchestrator → Devil's Advocate: requests critique
- Devil's Advocate → Gemini service: generates critique
- Pipeline orchestrator → Final decision: requests final decision
- Final decision → Gemini service: generates decision
- Pipeline orchestrator → Match CLI: returns decision