MCP App StoreMCP App Store
from naval-skill182
A Naval Ravikant–inspired cognitive operating system: role-play guidance, five mental models, heuristics, and an agentic workflow to answer career, leverage, an
Emulates a Naval Ravikant–style reasoning framework to answer questions about leverage, specific knowledge, wealth, and desire management. When activated, it instructs the agent to adopt Naval's voice, ask clarifying questions, run web research for fact-based queries, and respond with concise, model-driven guidance. The SKILL.md includes role rules, checkpoints, failure modes, decision heuristics, and a clear expression style to keep outputs consistent.
Use this skill when the user requests advice in Naval's perspective (phrases like 'Naval would say', 'use Naval's perspective', or specific Naval concepts such as 'specific knowledge', 'leverage', or 'desire audit'). Trigger only for topics tied to Naval's core ideas—leverage, career choices, wealth-building frameworks—not for generic decision-making requests.
Works with conversational agents capable of web search and tool use (to perform required research when questions need up-to-date facts). Best suited for agents that can preserve role state and follow strict output style constraints.
Naval Perspective is a role-play skill that activates Naval Ravikant's cognitive operating system — five mental models, eight decision heuristics, and an agentic workflow for research-backed answers. No scripts were included; the skill is purely prompt-based. The SKILL.md is exceptionally detailed with clear step protocols, checkpoints, and failure-mode fallbacks, but the description and much of the content is in Chinese, limiting broad accessibility. Architecture is decent but lacks bundled references/ or scripts/ directories despite referencing them.
Well-crafted role-play/prompt-engineering skill. No security concerns whatsoever — no scripts, no network calls beyond expected web search, no data exfiltration. The agentic protocol (research-then-answer) is a thoughtful pattern. Main limitation is language: primarily Chinese with some English quotes. Architecture could improve by bundling the referenced materials.
Argue — Multi-Agent Debate Engine
Run structured multi-agent debates and cross-examinations to reach higher-confidence consensus for strategic decisions and risk analysis.
Tool
Framework for finding, evaluating, and integrating software tools and building workflows that reduce switching costs and improve productivity.
Altitude & Horizon Framework
Diagnose and coach the PM→Director transition using a two-axis mental model (Altitude: scope, Horizon: time) and practical tools like the Cascading Context Map.
Pig Skill
Create a personalised conversational persona from QQ group chat records or manual inputs; generates a reusable Skill that speaks like that person and supports i
Elon-Plan — The Algorithm Planner
Run 'The Algorithm' to produce aggressive, first-principles project plans: question requirements, delete, simplify, accelerate, then automate — with owners and
Structured Product Debate
Run a multi-agent, cross-model adversarial debate to produce a high-quality product requirements document (PRD) and decision log.
Fear (Pre-mortem & Risk Amplifier)
Structured pre-mortem generator that amplifies plausible worst-case scenarios, ranks failure modes, and produces concrete mitigation recommendations.
UX Audit (v2)
Audit screens and flows against Nielsen's heuristics and mobile UX best practices (StyleSeed Toss context), preserving upstream provenance and copied workflow f