Game Theory Strategist

• Outcome: Pass
• Latest audit: May 28, 2026 at 3:46 AM UTC
• Version:
  1.0.1

Overview

This is a coherent game-theory coaching skill with local-only helper scripts, but users should treat its negotiation advice as heuristic and potentially sensitive in personal or high-stakes disputes.

Install only if you want a strategic-analysis lens for negotiations and conflicts. Treat payoff numbers, equilibrium labels, and action plans as illustrative, not legal, financial, therapeutic, or professional advice. Be especially cautious in family, employment, legal, financial, or vulnerable-person contexts, and do not use the skill to pressure, isolate, deceive, or coerce another person.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The README describes very broad triggering conditions such as everyday conflicts, negotiations, and situations involving uncertainty, which can cause the skill to activate in many ordinary conversations. Over-broad invocation increases the chance of inappropriate routing and overconfident strategic framing in contexts that may require legal, financial, interpersonal, or mental-health-sensitive judgment rather than a simplified game-theory model.

Missing User Warnings

Low

• Confidence: 76% confidence
• Finding: The README explicitly notes that an output warning about qualitative utility estimates is still missing, meaning users may see payoff matrices and recommendations without a sufficiently prominent caveat that the numbers are illustrative only. In a strategy or negotiation skill, this can mislead users into treating speculative utilities and equilibrium labels as rigorous decision support, increasing the risk of poor real-world decisions.

Vague Triggers

High

• Confidence: 94% confidence
• Finding: The trigger scope is extremely broad, covering many common life, work, and conflict scenarios, which can cause unintended invocation in contexts where specialized strategic framing is inappropriate. Over-broad activation increases the chance of the skill steering unrelated conversations, producing manipulative or misaligned advice, and overshadowing safer or more relevant skills.

Natural-Language Policy Violations

Medium

• Confidence: 83% confidence
• Finding: Forcing output sections into Spanish without user opt-in can cause misunderstanding of strategic advice, especially in negotiations or conflict scenarios where nuance matters. While not a classic security flaw, it can degrade user comprehension and lead to unsafe decisions based on misunderstood recommendations.

Natural-Language Policy Violations

Medium

• Confidence: 90% confidence
• Finding: The referenced section explicitly recommends brinkmanship tactics such as making threats more credible by escalating shared risk and 'burning bridges' through irreversible commitment devices. In a skill positioned for negotiation, conflict, and personal disputes, this can steer users toward coercive, manipulative, or escalatory behavior without safeguards, consent boundaries, or screening for high-risk contexts.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The suggestion to "lower the outside option via timing/urgency" explicitly recommends changing another party's alternatives or decision context to make refusal less attractive. In a negotiation/game-theory skill focused on real interpersonal and business conflicts, this normalizes coercive or manipulative pressure tactics without any ethical guardrails, which can facilitate unfair influence over vulnerable users or counterparties.

VirusTotal

VirusTotal findings are pending for this skill version.