---
name: yao-gametheory-skill
description: Route CEO-facing competition, negotiation, channel, pricing, platform, M&A, financing, competitor-response, alliance, and regulator problems through an applied game-theory framework router when decisions depend on opponent reactions, credible commitments, signals, payoff/equilibrium logic, historical behavior, dynamic updates, or Markdown/HTML/Word/PDF report export. Do not use for textbook tutoring, proofs, gambling, generic brainstorming, or final legal/financial/regulatory advice.
---

# Yao Game Theory Skill

## Use This Skill For

- convert CEO-facing strategic interactions into game theory reports
- choose and combine game theory frameworks from `references/framework-catalog.md`
- answer "对手可能怎么反应" and "我们的承诺动作是否可信"
- use historical behavior and reference-class experience to adjust rationality and commitment probabilities
- update prior reports when a player makes a new move
- export synchronized Markdown, HTML, Word/DOCX, and PDF reports

## Do Not Route Here

- game theory homework, theorem proofs, or textbook explanation with no strategic report
- generic strategy brainstorming where no player can react strategically
- final legal, financial, investment, antitrust, or regulatory advice
- manipulative, deceptive, coercive, or unlawful competitive tactics

## Default Workflow

1. Convert the request with `references/intake-contract.md`; use `references/dynamic-iteration-loop.md` for incomplete input or later player moves.
2. Route through `references/framework-catalog.md`: select one primary framework and 2-4 secondary lenses.
3. Instantiate the combination with `references/game-model-playbook.md`: players, strategies, payoffs, timing, beliefs, and equilibrium logic.
4. Calibrate player rationality with real historical behavior and experience references before trusting payoff-matrix inference.
5. Test commitments and signals with `references/commitment-signal-checklist.md`.
6. Run `references/strategic-hygiene-checklist.md`, `references/strategy-readiness-loop.md`, and `references/sensitivity-and-safety.md`.
7. Build framework selection, opponent reactions, payoff matrix, historical behavior calibration, readiness, sensitivity, next information, and update triggers.
8. Run `scripts/generate_report_bundle.py`; finalize with `references/strategic-report-contract.md` and `references/report-export-pipeline.md`.

## Output Contract

- Produce an executive strategy report, not a theory lecture.
- Put the recommendation, opponent reaction forecast, and credible-commitment call before technical sections.
- Mark payoffs, probabilities, and assumptions as observed, estimated, or assumed.
- Include frameworks, payoff matrix, reactions, equilibria, historical rationality calibration, commitment/signal checks, readiness, sensitivity, next information, update log, triggers, and caveats.
- Default language is Simplified Chinese; generate PDF automatically when possible, otherwise keep HTML print-ready.

## Reference Map

- `references/intake-contract.md`: request-to-game-brief conversion
- `references/framework-catalog.md`: game theory framework library and AI application router
- `references/game-model-playbook.md`: players, strategies, payoffs, timing, equilibrium reasoning
- `references/commitment-signal-checklist.md`: credibility and signal-quality checks
- `references/dynamic-iteration-loop.md`: multi-turn and opponent-update loop
- `references/strategy-readiness-loop.md`: readiness score and remaining-gap loop
- `references/strategic-hygiene-checklist.md`: strategic assumption hygiene checks
- `references/sensitivity-and-safety.md`: stress tests and safety boundaries
- `references/strategic-report-contract.md`: required report sections and schema alignment
- `references/report-export-pipeline.md`: artifact generation and print/PDF rules
- `scripts/generate_report_bundle.py`: report builder and exporter
- `templates/report.schema.json`: structured report schema
- `evals/trigger_cases.json`: route positives, negatives, and near-neighbor cases