legends-mcp

#!/usr/bin/env node /** * Example: Resolving Conflicting Advice from Multiple Legends * * This example demonstrates: * 1. Summoning multiple legends for the same strategic decision * 2. Comparing their distinct mental models and frameworks * 3. Identifying conflicts and areas of agreement * 4. Synthesizing a balanced decision matrix * 5. Using MCP's parallel tool calling for efficient comparison */ import { Client } from '@modelcontextprotocol/sdk/client/index.js'; import { StdioClientTransport } from '@modelcontextprotocol/sdk/client/stdio.js'; // Initialize MCP client const transport = new StdioClientTransport({ command: 'npx', args: ['legends-mcp'], }); const client = new Client( { name: 'conflict-resolution-example', version: '1.0.0' }, { capabilities: {} } ); await client.connect(transport); console.log('='.repeat(70)); console.log('Conflicting Advice Resolution: Strategic Decision Analysis'); console.log('='.repeat(70)); console.log(); /** * Strategic Decision Scenario */ const decision = { context: "Should we pivot from B2B SaaS to B2C marketplace?", background: ` - Current: B2B SaaS with 50 paying customers ($500k ARR) - Traction: Growing 10% MoM, but slow enterprise sales cycle - Opportunity: Noticed organic B2C demand (200+ signups/week) - Team: 5 engineers, 1 PM, 1 sales - Runway: 18 months `.trim(), stakeholders: ['investors', 'team', 'customers'], }; /** * Legend Perspectives to Compare */ const legendPerspectives = [ { id: 'paul-graham', expertise: 'Startup fundamentals & pivoting' }, { id: 'peter-thiel', expertise: 'Zero-to-one thinking & monopoly' }, { id: 'reid-hoffman', expertise: 'Blitzscaling & network effects' }, { id: 'jeff-bezos', expertise: 'Customer obsession & long-term thinking' }, ]; /** * Data Structure for Advice Comparison */ class AdviceComparison { constructor() { this.perspectives = []; this.conflictAreas = []; this.agreements = []; this.decisionMatrix = { factors: [], legendScores: {}, }; } addPerspective(legendId, legendName, advice) { this.perspectives.push({ legendId, legendName, advice }); } analyze() { // Extract key themes from each perspective const themes = {}; this.perspectives.forEach(({ legendId, legendName, advice }) => { // Extract themes from principles and patterns const legendThemes = this.extractThemes(advice); themes[legendId] = legendThemes; }); // Find conflicts (themes where legends disagree) this.findConflicts(themes); // Find agreements (common themes across legends) this.findAgreements(themes); // Build decision matrix this.buildDecisionMatrix(); } extractThemes(advice) { return { speed: advice.principles.some(p => p.toLowerCase().includes('fast') || p.toLowerCase().includes('move')), risk: advice.principles.some(p => p.toLowerCase().includes('risk') || p.toLowerCase().includes('bet')), focus: advice.principles.some(p => p.toLowerCase().includes('focus') || p.toLowerCase().includes('one thing')), scale: advice.principles.some(p => p.toLowerCase().includes('scale') || p.toLowerCase().includes('growth')), customer: advice.principles.some(p => p.toLowerCase().includes('customer') || p.toLowerCase().includes('user')), }; } findConflicts(themes) { const themeKeys = ['speed', 'risk', 'focus', 'scale', 'customer']; themeKeys.forEach(theme => { const values = Object.entries(themes).map(([id, t]) => ({ id, value: t[theme] })); const hasConflict = values.some(v => v.value) && values.some(v => !v.value); if (hasConflict) { this.conflictAreas.push({ theme, for: values.filter(v => v.value).map(v => v.id), against: values.filter(v => !v.value).map(v => v.id), }); } }); } findAgreements(themes) { const themeKeys = ['speed', 'risk', 'focus', 'scale', 'customer']; themeKeys.forEach(theme => { const values = Object.values(themes).map(t => t[theme]); const allAgree = values.every(v => v === values[0]); if (allAgree && values[0]) { this.agreements.push(theme); } }); } buildDecisionMatrix() { // Define decision factors this.decisionMatrix.factors = [ { name: 'Product-Market Fit', weight: 0.3 }, { name: 'Growth Potential', weight: 0.25 }, { name: 'Execution Risk', weight: 0.2 }, { name: 'Competitive Advantage', weight: 0.15 }, { name: 'Resource Fit', weight: 0.1 }, ]; } generateSummary() { return { totalPerspectives: this.perspectives.length, conflictAreas: this.conflictAreas, agreements: this.agreements, recommendation: this.synthesizeRecommendation(), }; } synthesizeRecommendation() { // Weight agreement areas higher if (this.agreements.length > this.conflictAreas.length) { return 'Strong consensus among legends - proceed with confidence'; } else if (this.conflictAreas.length > this.agreements.length) { return 'Significant disagreement - deeper analysis needed before decision'; } else { return 'Mixed signals - consider phased approach or additional validation'; } } } /** * Example 1: Summon Multiple Legends in Parallel */ async function gatherMultiplePerspectives() { console.log('📞 Summoning Multiple Legends in Parallel\n'); console.log(`Decision: ${decision.context}\n`); try { // Parallel summon for efficiency const summonPromises = legendPerspectives.map(legend => client.callTool({ name: 'summon_legend', arguments: { legend_id: legend.id }, }).then(result => ({ id: legend.id, data: JSON.parse(result.content[0].text), })) ); const results = await Promise.all(summonPromises); console.log(`✅ Summoned ${results.length} legends:\n`); results.forEach(({ id, data }) => { console.log(` • ${data.name} - ${data.quick_ref.voice.tone}`); }); console.log(); return results; } catch (error) { console.error('❌ Error summoning legends:', error.message); throw error; } } /** * Example 2: Extract and Compare Mental Models */ async function compareMentalModels(legends) { console.log('🧠 Comparing Mental Models\n'); const comparison = new AdviceComparison(); legends.forEach(({ id, data }) => { comparison.addPerspective(id, data.name, data.quick_ref); console.log(`${data.name}'s Framework:`); console.log(` Core Principle: ${data.quick_ref.principles[0]}`); console.log(` Key Pattern: ${data.quick_ref.patterns[0].name}`); console.log(` Never Says: "${data.quick_ref.never_say[0]}"`); console.log(); }); // Analyze for conflicts and agreements comparison.analyze(); return comparison; } /** * Example 3: Identify Conflict Areas */ function displayConflicts(comparison) { console.log('⚔️ Conflict Areas\n'); if (comparison.conflictAreas.length === 0) { console.log(' No major conflicts detected - legends are aligned!\n'); return; } comparison.conflictAreas.forEach(conflict => { console.log(` Theme: ${conflict.theme.toUpperCase()}`); console.log(` Supporting: ${conflict.for.join(', ')}`); console.log(` Cautionary: ${conflict.against.join(', ')}`); console.log(); }); } /** * Example 4: Find Areas of Agreement */ function displayAgreements(comparison) { console.log('🤝 Areas of Agreement\n'); if (comparison.agreements.length === 0) { console.log(' No universal agreement - legends have diverse views\n'); return; } console.log(` All ${comparison.perspectives.length} legends agree on:`); comparison.agreements.forEach(theme => { console.log(` • ${theme.toUpperCase()}`); }); console.log(); } /** * Example 5: Build Decision Matrix */ async function buildDecisionMatrix(legends) { console.log('📊 Decision Matrix: B2B SaaS vs B2C Marketplace\n'); const factors = [ { name: 'Product-Market Fit', b2b: 7, b2c: 5 }, { name: 'Growth Potential', b2b: 6, b2c: 9 }, { name: 'Execution Risk', b2b: 3, b2c: 8 }, { name: 'Competitive Advantage', b2b: 7, b2c: 4 }, { name: 'Resource Fit', b2b: 8, b2c: 5 }, ]; console.log(' Factor | B2B | B2C | Gap'); console.log(' ' + '-'.repeat(50)); let b2bTotal = 0; let b2cTotal = 0; factors.forEach(factor => { const gap = factor.b2b - factor.b2c; const arrow = gap > 0 ? '←' : gap < 0 ? '→' : '='; console.log(` ${factor.name.padEnd(25)} | ${factor.b2b}/10 | ${factor.b2c}/10 | ${arrow} ${Math.abs(gap)}`); b2bTotal += factor.b2b; b2cTotal += factor.b2c; }); console.log(' ' + '-'.repeat(50)); console.log(` TOTALS | ${b2bTotal}/50 | ${b2cTotal}/50 |`); console.log(); // Legend-weighted scores console.log(' Legend-Weighted Recommendation:\n'); legends.forEach(({ id, data }) => { const recommendation = getRecommendation(id, factors); console.log(` ${data.name}: ${recommendation}`); }); console.log(); } function getRecommendation(legendId, factors) { // Different legends weight factors differently const weights = { 'paul-graham': { pmf: 0.5, growth: 0.2, risk: 0.15, advantage: 0.1, resource: 0.05 }, 'peter-thiel': { pmf: 0.3, growth: 0.15, risk: 0.2, advantage: 0.3, resource: 0.05 }, 'reid-hoffman': { pmf: 0.2, growth: 0.4, risk: 0.15, advantage: 0.15, resource: 0.1 }, 'jeff-bezos': { pmf: 0.4, growth: 0.2, risk: 0.1, advantage: 0.2, resource: 0.1 }, }; const weight = weights[legendId]; const b2bScore = (factors[0].b2b * weight.pmf) + (factors[1].b2b * weight.growth) + (factors[2].b2b * weight.risk) + (factors[3].b2b * weight.advantage) + (factors[4].b2b * weight.resource); const b2cScore = (factors[0].b2c * weight.pmf) + (factors[1].b2c * weight.growth) + (factors[2].b2c * weight.risk) + (factors[3].b2c * weight.advantage) + (factors[4].b2c * weight.resource); if (b2bScore > b2cScore) { return `Stay B2B (Score: ${b2bScore.toFixed(1)} vs ${b2cScore.toFixed(1)})`; } else if (b2cScore > b2bScore) { return `Pivot to B2C (Score: ${b2cScore.toFixed(1)} vs ${b2bScore.toFixed(1)})`; } else { return `Hybrid approach (Scores tied: ${b2bScore.toFixed(1)})`; } } /** * Example 6: Synthesize Final Recommendation */ function synthesizeFinalRecommendation(comparison) { console.log('💡 Synthesized Recommendation\n'); const summary = comparison.generateSummary(); console.log(` Analyzed ${summary.totalPerspectives} legendary perspectives`); console.log(` Conflicts detected: ${summary.conflictAreas.length}`); console.log(` Agreements found: ${summary.agreements.length}`); console.log(); console.log(` Overall Assessment:`); console.log(` ${summary.recommendation}`); console.log(); // Provide actionable next steps console.log(' Next Steps:'); if (summary.conflictAreas.length > 0) { console.log(' 1. Deep dive into conflict areas with targeted legend insights'); console.log(' 2. Run experiments to validate conflicting hypotheses'); console.log(' 3. Weight legend advice by relevance to your specific context'); } if (summary.agreements.length > 0) { console.log(' 1. Leverage areas of agreement as strong signals'); console.log(' 2. Build execution plan around consensus themes'); console.log(' 3. Communicate unified legend wisdom to stakeholders'); } console.log(); } /** * Main execution */ async function main() { try { console.log('Strategic Decision Context:'); console.log(decision.background); console.log(); console.log('─'.repeat(70) + '\n'); // Step 1: Gather multiple perspectives in parallel const legends = await gatherMultiplePerspectives(); console.log('─'.repeat(70) + '\n'); // Step 2: Compare mental models const comparison = await compareMentalModels(legends); console.log('─'.repeat(70) + '\n'); // Step 3: Display conflicts displayConflicts(comparison); console.log('─'.repeat(70) + '\n'); // Step 4: Display agreements displayAgreements(comparison); console.log('─'.repeat(70) + '\n'); // Step 5: Build decision matrix await buildDecisionMatrix(legends); console.log('─'.repeat(70) + '\n'); // Step 6: Synthesize final recommendation synthesizeFinalRecommendation(comparison); console.log('🎯 MCP Integration Benefits:'); console.log(' • Parallel tool calling for efficient legend summoning'); console.log(' • Structured mental model comparison via YAML schemas'); console.log(' • Conflict detection through principle/pattern analysis'); console.log(' • Decision matrix generation from distinct frameworks'); console.log(); } catch (error) { console.error('Fatal error:', error); } finally { await client.close(); } } main().catch(console.error);