mcorbett51090/ravenclaude/game-theory-basics
Claude Code and agent skill by mcorbett51090/ravenclaude
- community
Checking that the SKILL.md file is still in the repository.
Use this when multiple agents or parties have incentives — sketch payoffs, dominant strategies, and equilibrium intuition for coordination, gates, and escalation.
Install
npx vskill@latest install mcorbett51090/ravenclaude/game-theory-basicsSKILL.md
The same SKILL.md was published earlier as mcorbett51090/ravenclaude/game-theory-basics.
Game Theory Basics
Companion to RavenClaude Core and Quantitative Problem Solving.
Stance
- Payoffs first — who gets what under each outcome.
- Dominant strategy if it exists; else mutual best replies / equilibrium intuition.
- Not everything is a game — if the opponent is nature or a bug, use Quantitative Problem Solving instead.
- Cooperate by default inside the team — hiding failure or bypassing CoS/gates is negative-sum for Matthew.
Workflow 1 — Normal-form sketch
Players, actions, payoffs (ordinal is fine). Classic Prisoner's Dilemma when Temptation > Reward > Punishment > Sucker. Inside Matthew's bot team, set long-run payoffs so honest escalate / share-evidence wins.
Workflow 2 — Checklist
- Who are the players?
- What moves exist this turn?
- Payoffs for Matthew first, then each player
- Is there a dominant strategy?
- If simultaneous: reasonable Nash / mutual best reply?
- If sequential: backward induction from the last node
- Prefer strategies incentive-compatible with wall escalation and gates
Workflow 3 — Grok Bot mappings
- Hide wall vs escalate to CoS: cooperate = escalate with evidence
- Two bots disagree: forge panels + critic, not volume war
- Money/delete approval: do not bypass; waiting is the move
- API rate limits: repeated game — backoff sustains access
Deliverable shape
Players + moves; payoff notes; recommended strategy in one sentence; what would change it.
Anti-patterns
Treating a bug as an adversary when nature models suffice; defecting against process to save tokens; assuming zero-sum when a correct fix grows the pie.
Credit
Lightweight companion for RavenClaude Core multi-agent / human loops.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingplugins/ravenclaude-core/skills/game-theory-basics/SKILL.md
Popularity
- Repository stars
- 7
- Repository forks
- 1
- 7-day trend score
- 15
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/mcorbett51090/ravenclaude/game-theory-basics)