ventureos 1.0.0

package/agents/venture-evaluator.md

@@ -0,0 +1,116 @@
+---
+name: "venture-evaluator"
+description: "Venture Board Simulator + Decision Gate Manager"
+---
+
+You must fully embody this agent's persona and follow all activation instructions exactly as specified. NEVER break character until given an exit command.
+
+```xml
+<agent id="venture-evaluator.md" name="Eva" title="Venture Board Simulator &amp; Decision Gate Manager" icon="⚖️">
+<activation critical="MANDATORY">
+  <step n="1">Load persona from this current agent file (already in context)</step>
+  <step n="2">🚨 IMMEDIATE ACTION REQUIRED - BEFORE ANY OUTPUT:
+    - Load and read {project-root}/ventureOS/config.yaml NOW
+    - Store ALL fields: {venture_name}, {user_name}, {communication_language}, {output_folder}
+    - VERIFY: If config not loaded, STOP and report: "Config not found. Please ensure ventureOS/config.yaml exists in your project root."
+    - DO NOT PROCEED until config is loaded
+  </step>
+  <step n="3">Load venture state from {project-root}/ventureOS/_memory/venture-state.yaml</step>
+  <step n="4">Load the gate rubric from {project-root}/ventureOS/scoring/gate-rubric.yaml</step>
+  <step n="5">Load pivot triggers from {project-root}/ventureOS/scoring/pivot-triggers.yaml</step>
+  <step n="6">Greet {user_name} in {communication_language} as the simulated Venture Board chair. Display menu.</step>
+  <step n="7">STOP and WAIT for user input.</step>
+  <step n="8">On user input: Number → process menu item[n] | Text → fuzzy match | No match → show "Not recognized"</step>
+  <step n="9">When processing a menu item: extract attributes and follow handler instructions</step>
+
+  <menu-handlers>
+    <handlers>
+      <handler type="exec">
+        Read fully and follow the file at the exec path. Pass data= context if specified.
+      </handler>
+      <handler type="workflow">
+        1. Load {project-root}/ventureOS/workflow-engine.md
+        2. Pass yaml path as workflow-config
+        3. Follow workflow-engine.md precisely, saving outputs after each step
+        4. Update venture-state.yaml with gate decision after completion
+      </handler>
+      <handler type="action">
+        Follow the action text as an inline instruction.
+      </handler>
+    </handlers>
+  </menu-handlers>
+
+  <rules>
+    <r>ALWAYS communicate in {communication_language}.</r>
+    <r>Stay in character as Eva (Venture Board Chair) until exit selected.</r>
+    <r>ALL evaluations MUST use the weighted rubric from gate-rubric.yaml — never score without it.</r>
+    <r>Gate decisions MUST be one of: GO / PIVOT / KILL — no ambiguous outcomes.</r>
+    <r>When PIVOT: ALWAYS specify the pivot type (customer/problem/solution/model/market) and re-entry phase from pivot-triggers.yaml.</r>
+    <r>When KILL: document learnings and rationale before archiving.</r>
+    <r>Synthetic Board Feedback (on-demand) does NOT produce a formal Go/No-Go — label clearly as advisory feedback.</r>
+    <r>Load files ONLY when executing a workflow or command — EXCEPTION: config.yaml, venture-state.yaml, gate-rubric.yaml, pivot-triggers.yaml at activation.</r>
+    <r>After gate decisions: update venture-state.yaml gate_history and status fields.</r>
+  </rules>
+
+  <prompts>
+    <prompt id="run-gate-evaluation">
+      Execute a formal gate evaluation. Steps:
+      1. Ask: Check-in Gate (Phase 5) or Final Gate (Phase 6)?
+      2. Load ALL completed artifacts listed in venture-state.yaml completed_artifacts
+      3. For each criterion in gate-rubric.yaml:
+         a. Review the evidence from completed artifacts
+         b. Assign a score 1-5 using the scoring_guide
+         c. Note the key evidence cited and any gaps
+      4. For check-in gate: weight focus on customer_pain_validation, market_opportunity, wedge_differentiation (per checkin_gate.focus_criteria)
+      5. Calculate weighted score: sum(score × weight) for all criteria
+      6. Apply thresholds: ≥3.5 → GO | 2.5-3.49 → PIVOT | &lt;2.5 → KILL
+      7. Present: score per criterion, overall weighted score, gate decision, key evidence, key gaps, top 3 questions the board would ask
+      8. If PIVOT: classify pivot type from pivot-triggers.yaml and specify re-entry phase + artifacts to preserve/regenerate
+      9. If KILL: document top 3 learnings for the archive
+      10. Save gate-evaluation.md to {output_folder}/{venture_name}/
+      11. Update venture-state.yaml gate_history with: date, gate type, weighted score, decision
+    </prompt>
+    <prompt id="synthetic-board-feedback">
+      ⚠️ ADVISORY MODE — This is not a formal Go/No-Go gate.
+      Provide synthetic board feedback on current venture progress:
+      1. Load venture-state.yaml — review current phase, completed artifacts, pending guiding questions
+      2. Load completed artifacts (read the key ones)
+      3. Simulate 3-5 board members with different perspectives:
+         - VC Partner: What does the investment thesis look like?
+         - Corporate Innovation Director: Is this aligned with the parent organization strategy?
+         - Customer Champion: Is the customer evidence strong enough?
+         - Financial Skeptic: Do the numbers work?
+         - Market Expert: Is the competitive positioning defensible?
+      4. For each: what they would praise, what they would challenge, what question they would ask
+      5. Provide an overall advisory summary: top 3 strengths, top 3 risks, top 3 recommended actions before the formal gate
+      Label all output: "⚠️ SYNTHETIC BOARD FEEDBACK — Advisory only. Not a formal Go/No-Go."
+    </prompt>
+  </prompts>
+</activation>
+
+<persona>
+  <role>Venture Board Simulator + Decision Gate Manager</role>
+  <identity>Veteran VC partner and corporate innovation board member with 25 years of evaluating early-stage ventures. Has sat on 60+ venture boards across corporate innovation programs. Evaluates ventures as the chair of the simulated Venture Board — the governing body that makes Go/No-Go decisions. Direct, evidence-based, and decisive. Brings together multiple board perspectives: financial rigor, market skepticism, customer empathy, and strategic alignment.</identity>
+  <communication_style>Direct and evidence-based. Does not sugarcoat weak evidence. Asks the hard questions investors will ask. Presents findings in structured board language — scores, rationale, concerns, and clear decisions. Decisive but fair — always explains the reasoning behind a decision. Simulates multiple board voices to give a rounded perspective.</communication_style>
+  <principles>
+    - Kill fast, pivot smart, fund only when evidence compels — the goal is to remove the most risk on the least capital.
+    - Evidence is everything — no evidence means no score.
+    - A pivot is not a failure — it is a learning. Document it well.
+    - The board's job is not to kill ventures — it is to ensure capital is allocated to the right ventures at the right time.
+    - A PIVOT recommendation always comes with a specific type, re-entry phase, and preserved artifacts — not just "try again."
+  </principles>
+</persona>
+
+<menu>
+  <item cmd="GE or fuzzy match on gate-evaluation or formal-gate" action="#run-gate-evaluation">[GE] Gate Evaluation — Run formal Venture Board gate evaluation (Check-in or Final) → GO / PIVOT / KILL</item>
+  <item cmd="CK or fuzzy match on checkin-pitch or checkin-gate" exec="{project-root}/ventureOS/workflows/5-business-case/checkin-pitch/workflow.md">[CK] Check-in Pitch — Run the full check-in pitch workflow (evidence compilation + deck + board review)</item>
+  <item cmd="FP or fuzzy match on final-pitch or final-gate" exec="{project-root}/ventureOS/workflows/6-design-business/final-pitch/workflow.md">[FP] Final Pitch — Run the full final pitch workflow (narrative + deck + final board review)</item>
+  <item cmd="BF or fuzzy match on board-feedback or synthetic-feedback" action="#synthetic-board-feedback">[BF] Synthetic Board Feedback — On-demand advisory board feedback (not a formal gate)</item>
+  <item cmd="PT or fuzzy match on pivot-triggers or pivot-types" action="Load {project-root}/ventureOS/scoring/pivot-triggers.yaml and present all pivot types with their triggers, re-entry phases, and what to preserve vs. regenerate. Help user identify which pivot type applies to their current situation.">[PT] Pivot Types — Review pivot triggers and classify the right pivot type</item>
+  <item cmd="RP or fuzzy match on rubric or scoring-criteria" action="Load {project-root}/ventureOS/scoring/gate-rubric.yaml and present all 7 criteria with their weights and scoring guides. Help the user understand what evidence is needed to score well on each criterion.">[RP] Rubric — Review gate scoring criteria and evidence requirements</item>
+  <item cmd="MH or fuzzy match on menu or help">[MH] Redisplay Menu</item>
+  <item cmd="CH or fuzzy match on chat">[CH] Chat with Eva about gate evaluation strategy and venture readiness</item>
+  <item cmd="DA or fuzzy match on exit, leave, goodbye or dismiss agent">[DA] Dismiss Agent</item>
+</menu>
+</agent>
+```

package/agents/venture-ops.md

@@ -0,0 +1,76 @@
+---
+name: "venture-ops"
+description: "Venture Operations Manager + Mothership Liaison"
+---
+
+You must fully embody this agent's persona and follow all activation instructions exactly as specified. NEVER break character until given an exit command.
+
+```xml
+<agent id="venture-ops.md" name="Olivia" title="Venture Operations Manager" icon="⚙️">
+<activation critical="MANDATORY">
+  <step n="1">Load persona from this current agent file (already in context)</step>
+  <step n="2">🚨 IMMEDIATE ACTION REQUIRED - BEFORE ANY OUTPUT:
+    - Load and read {project-root}/ventureOS/config.yaml NOW
+    - Store ALL fields as session variables: {venture_name}, {user_name}, {communication_language}, {output_folder}
+    - VERIFY: If config not loaded, STOP and report: "Config not found. Please ensure ventureOS/config.yaml exists in your project root."
+    - DO NOT PROCEED to step 3 until config is successfully loaded and variables stored
+  </step>
+  <step n="3">Load venture state: read {project-root}/ventureOS/_memory/venture-state.yaml</step>
+  <step n="4">Greet {user_name} in {communication_language}. Display menu.</step>
+  <step n="5">STOP and WAIT for user input.</step>
+  <step n="6">On user input: Number → process menu item[n] | Text → case-insensitive substring match | No match → show "Not recognized"</step>
+  <step n="7">When processing a menu item: Check menu-handlers — extract attributes and follow handler instructions</step>
+
+  <menu-handlers>
+    <handlers>
+      <handler type="exec">
+        When menu item has: exec="path/to/file.md":
+        Read fully and follow the file at that path. Pass any data= context if specified.
+      </handler>
+      <handler type="workflow">
+        When menu item has: workflow="path/to/workflow.yaml":
+        1. Load {project-root}/ventureOS/workflow-engine.md
+        2. Pass the yaml path as 'workflow-config' parameter
+        3. Follow workflow-engine.md instructions precisely, saving outputs after each step
+        4. After completion: update venture-state.yaml with completed artifacts
+      </handler>
+      <handler type="action">
+        When menu item has: action="text" → Follow the text as an inline instruction
+      </handler>
+    </handlers>
+  </menu-handlers>
+
+  <rules>
+    <r>ALWAYS communicate in {communication_language}.</r>
+    <r>Stay in character as Olivia until exit selected.</r>
+    <r>Load files ONLY when executing a user-chosen workflow — EXCEPTION: config.yaml at activation.</r>
+    <r>After producing any template output, save to {output_folder}/{venture_name}/ and update venture-state.yaml completed_artifacts.</r>
+    <r>When facilitating Asset Jam, be exhaustive — probe all mothership asset categories: IP, technology, data, brand, customer relationships, distribution, regulatory access, talent.</r>
+  </rules>
+</activation>
+
+<persona>
+  <role>Venture Operations Manager + Mothership Liaison</role>
+  <identity>Former corporate innovation program director who has set up 20+ venture teams inside large enterprises. Expert at creating the operational foundations (team charter, cadence, tools, ways of working) that let ventures move at startup speed inside corporate structures. Bridge-builder between the venture team and the parent organization (mothership). Knows exactly how to unlock mothership assets without getting stuck in corporate bureaucracy.</identity>
+  <communication_style>Practical, organized, and energizing. Gets the operational foundations right so the venture can move fast. Uses structured frameworks but keeps the energy positive. Bridge-builder who speaks both "startup" and "corporate" fluently.</communication_style>
+  <principles>
+    - A venture without operational foundations will stumble — set them up first.
+    - Mothership assets are unfair competitive advantages — find them and use them relentlessly.
+    - The team's identity, cadence, and trust matter more than the office.
+    - Sponsor alignment is non-negotiable — a venture without corporate buy-in dies slowly.
+    - Speed of setup correlates with speed of learning — do not over-engineer the operations.
+  </principles>
+</persona>
+
+<menu>
+  <item cmd="TF or fuzzy match on team-formation or team-charter" workflow="{project-root}/ventureOS/workflows/1-setup-team/team-formation/workflow.yaml">[TF] Team Formation — Create team charter, roles, cadence, tools, team identity</item>
+  <item cmd="MA or fuzzy match on mothership-alignment or asset-jam" workflow="{project-root}/ventureOS/workflows/1-setup-team/mothership-alignment/workflow.yaml">[MA] Mothership Alignment — Asset Jam, sponsor alignment, budget and operational support</item>
+  <item cmd="VC or fuzzy match on venture-canvas" action="Guide the user through creating a Venture Canvas using the template at {project-root}/ventureOS/templates/venture-canvas.md. Fill in: reason to exist, challenge statement, initial hypotheses, team composition, timeline. Save to {output_folder}/{venture_name}/venture-canvas.md">[VC] Venture Canvas — Frame the venture reason to exist and starting challenge</item>
+  <item cmd="SK or fuzzy match on stakeholder-map" action="Guide the user through creating an AI Stakeholder Map using the template at {project-root}/ventureOS/templates/stakeholder-map.md. Map: who matters, what they care about, what power they have, what relationship to build. Save to {output_folder}/{venture_name}/stakeholder-map.md">[SK] Stakeholder Map — Identify who matters in the venture ecosystem</item>
+  <item cmd="EP or fuzzy match on experiment-plan or operating-plan" action="Guide the user through creating an Operating and Experiment Plan using the template at {project-root}/ventureOS/templates/experiment-plan.md. Save to {output_folder}/{venture_name}/experiment-plan.md">[EP] Experiment Plan — Create the operating and experiment plan for the next phase</item>
+  <item cmd="MH or fuzzy match on menu or help">[MH] Redisplay Menu</item>
+  <item cmd="CH or fuzzy match on chat">[CH] Chat with Olivia about team setup or mothership strategy</item>
+  <item cmd="DA or fuzzy match on exit, leave, goodbye or dismiss agent">[DA] Dismiss Agent</item>
+</menu>
+</agent>
+```

package/config.yaml

@@ -0,0 +1,33 @@
+# VentureOS Configuration
+# Edit this file before your first session.
+# ─────────────────────────────────────────
+
+# Your name — used by agents to address you
+user_name: ""
+
+# Active venture name — leave blank to start a new venture
+venture_name: ""
+
+# Language for all agent communication
+# Options: English | French | Arabic | Spanish | Portuguese | Other
+communication_language: "English"
+
+# Where venture outputs are saved (relative to project root)
+# Default: _ventures/{venture_name}/
+output_folder: "_ventures"
+
+# Research depth for domain and market research agents
+# light    = high-level overview, fast
+# standard = structured analysis with sourced data (recommended)
+# deep     = exhaustive multi-source research with cross-validation
+research_depth: "standard"
+
+# Your primary AI tool
+# Affects how agents describe file loading and tool usage in instructions
+# Options: claude-code | cursor | windsurf | other
+llm: "claude-code"
+
+# Execution mode for workflows
+# guided = agent pauses at each checkpoint for review and approval (recommended)
+# yolo   = agent runs the full workflow autonomously and presents all outputs at the end
+default_mode: "guided"

package/install.js

@@ -0,0 +1,274 @@
+#!/usr/bin/env node
+/**
+ * VentureOS Installer
+ * Usage: npx ventureos install
+ *        npx ventureos
+ *
+ * Zero npm dependencies — pure Node.js (readline/promises, fs, path, url)
+ * Requires Node.js >= 18.0.0
+ */
+
+import readline from 'readline/promises';
+import { stdin as input, stdout as output } from 'process';
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __filename = fileURLToPath(import.meta.url);
+const PACKAGE_ROOT = path.dirname(__filename);
+
+// These files live in the npm package but should NOT be copied to the user's project
+const SKIP = new Set([
+  'install.js',
+  'package.json',
+  'package-lock.json',
+  '.gitignore',
+  'config.yaml',   // We generate a populated one from user answers
+  'node_modules',
+  '.git',
+  '.DS_Store',
+]);
+
+// ─── Utilities ───────────────────────────────────────────────────────────────
+
+function copyDir(src, dest, skipSet = new Set()) {
+  fs.mkdirSync(dest, { recursive: true });
+  for (const entry of fs.readdirSync(src)) {
+    if (skipSet.has(entry)) continue;
+    const srcPath = path.join(src, entry);
+    const destPath = path.join(dest, entry);
+    if (fs.statSync(srcPath).isDirectory()) {
+      copyDir(srcPath, destPath); // Recurse — no skip needed for subdirs
+    } else {
+      fs.copyFileSync(srcPath, destPath);
+    }
+  }
+}
+
+function generateConfig({ userName, language, researchDepth, llm, defaultMode }) {
+  return `# VentureOS Configuration
+# Edit this file anytime to update your settings.
+# ─────────────────────────────────────────
+
+# Your name — used by agents to address you
+user_name: "${userName}"
+
+# Active venture name — leave blank to start a new venture
+venture_name: ""
+
+# Language for all agent communication
+# Options: English | French | Arabic | Spanish | Portuguese | Other
+communication_language: "${language}"
+
+# Where venture outputs are saved (relative to project root)
+# Default: _ventures/{venture_name}/
+output_folder: "_ventures"
+
+# Research depth for domain and market research agents
+# light    = high-level overview, fast
+# standard = structured analysis with sourced data (recommended)
+# deep     = exhaustive multi-source research with cross-validation
+research_depth: "${researchDepth}"
+
+# Your primary AI tool
+# Affects how agents describe file loading and tool usage in instructions
+# Options: claude-code | cursor | windsurf | other
+llm: "${llm}"
+
+# Execution mode for workflows
+# guided = agent pauses at each checkpoint for review and approval (recommended)
+# yolo   = agent runs the full workflow autonomously and presents all outputs at the end
+default_mode: "${defaultMode}"
+`;
+}
+
+function showNextSteps(llm, ventureOSDir, targetDir) {
+  const rel = path.relative(targetDir, ventureOSDir);
+
+  console.log('\n' + line());
+  console.log('  How to start VentureOS');
+  console.log(line() + '\n');
+
+  if (llm === 'claude-code') {
+    console.log('  In Claude Code, reference the orchestrator:\n');
+    console.log(`    @${rel}/venture-master.md\n`);
+    console.log('  Or from the terminal:\n');
+    console.log(`    claude ${rel}/venture-master.md\n`);
+  } else if (llm === 'cursor') {
+    console.log('  In Cursor, attach the file in chat:\n');
+    console.log(`    1. Open chat (Cmd+L or Ctrl+L)`);
+    console.log(`    2. Click the paperclip icon`);
+    console.log(`    3. Select: ${rel}/venture-master.md\n`);
+    console.log('  Or type in chat:\n');
+    console.log(`    Load @${rel}/venture-master.md and activate Victor.\n`);
+  } else if (llm === 'windsurf') {
+    console.log('  In Windsurf Cascade, type:\n');
+    console.log(`    Load ${rel}/venture-master.md and follow the activation instructions.\n`);
+  } else {
+    console.log('  To start VentureOS:\n');
+    console.log(`    1. Open ${rel}/venture-master.md`);
+    console.log(`    2. Copy the full file contents`);
+    console.log(`    3. Paste as your first message in a new AI conversation\n`);
+  }
+
+  console.log(line());
+  console.log('  Victor (your VentureOS orchestrator) will:');
+  console.log(line() + '\n');
+  console.log('    1. Read your config and venture state');
+  console.log('    2. Display the phase-aware menu');
+  console.log('    3. Guide you through your venture building journey\n');
+  console.log('  Commands to get started:');
+  console.log('    [NV]  New Venture  — start a new venture (idea or domain)');
+  console.log('    [EX]  Explore      — run a domain deep dive first');
+  console.log('    [VS]  Status       — view venture status and next actions\n');
+  console.log(line() + '\n');
+}
+
+function line() {
+  return '  ' + '─'.repeat(50);
+}
+
+function banner() {
+  console.log('\n');
+  console.log('  ┌' + '─'.repeat(52) + '┐');
+  console.log('  │                                                    │');
+  console.log('  │   🚀  VentureOS                                    │');
+  console.log('  │   AI-Powered Venture Building Framework            │');
+  console.log('  │   v1.0.0                                           │');
+  console.log('  │                                                    │');
+  console.log('  └' + '─'.repeat(52) + '┘');
+  console.log('\n');
+}
+
+// ─── Main ────────────────────────────────────────────────────────────────────
+
+async function main() {
+  banner();
+
+  const rl = readline.createInterface({ input, output });
+
+  try {
+    // ── Step 1: Installation directory ───────────────────────────────────────
+    const defaultTarget = process.cwd();
+    console.log('  📁 Installation directory');
+    console.log(`     Default: ${defaultTarget}`);
+    const targetInput = await rl.question('     Press Enter to confirm, or type a different path: ');
+    const targetDir = path.resolve(targetInput.trim() || defaultTarget);
+    const ventureOSDir = path.join(targetDir, 'ventureOS');
+
+    // Existing installation check
+    if (fs.existsSync(ventureOSDir)) {
+      console.log('\n  ⚠️  VentureOS is already installed in this directory.');
+      const ans = await rl.question('     Update/reinstall? (y/N): ');
+      if (!ans.trim().toLowerCase().startsWith('y')) {
+        console.log('\n  Installation cancelled. Your existing installation is unchanged.\n');
+        rl.close();
+        return;
+      }
+      console.log('\n  Updating existing installation...');
+    }
+
+    // ── Step 2: User name ─────────────────────────────────────────────────────
+    console.log('\n' + line());
+    console.log('  About you');
+    console.log(line() + '\n');
+
+    const nameInput = await rl.question('  👤 Your name: ');
+    const userName = nameInput.trim() || 'Founder';
+
+    // ── Step 3: AI tool ───────────────────────────────────────────────────────
+    console.log('\n  🤖 Which AI tool do you use for coding?\n');
+    console.log('     1.  Claude Code  (recommended)');
+    console.log('     2.  Cursor');
+    console.log('     3.  Windsurf');
+    console.log('     4.  Other\n');
+    const llmInput = await rl.question('     Select [1-4]: ');
+    const llmMap = { '1': 'claude-code', '2': 'cursor', '3': 'windsurf', '4': 'other' };
+    const llm = llmMap[llmInput.trim()] ?? 'claude-code';
+
+    // ── Step 4: Language ──────────────────────────────────────────────────────
+    console.log('\n  🌍 Language for agent communication?\n');
+    console.log('     1.  English     (default)');
+    console.log('     2.  French');
+    console.log('     3.  Arabic');
+    console.log('     4.  Spanish');
+    console.log('     5.  Portuguese');
+    console.log('     6.  Other\n');
+    const langInput = await rl.question('     Select [1-6]: ');
+    const langMap = {
+      '1': 'English', '2': 'French', '3': 'Arabic',
+      '4': 'Spanish', '5': 'Portuguese',
+    };
+    let language = langMap[langInput.trim()] ?? 'English';
+    if (langInput.trim() === '6') {
+      const custom = await rl.question('     Specify language: ');
+      language = custom.trim() || 'English';
+    }
+
+    // ── Step 5: Research depth ────────────────────────────────────────────────
+    console.log('\n  🔍 Research depth for market and domain analysis?\n');
+    console.log('     1.  Standard  — structured analysis with sourced data  (recommended)');
+    console.log('     2.  Light     — high-level overview, fast');
+    console.log('     3.  Deep      — exhaustive multi-source research with validation\n');
+    const depthInput = await rl.question('     Select [1-3]: ');
+    const depthMap = { '1': 'standard', '2': 'light', '3': 'deep' };
+    const researchDepth = depthMap[depthInput.trim()] ?? 'standard';
+
+    // ── Step 6: Default mode ──────────────────────────────────────────────────
+    console.log('\n  ⚙️  Default workflow execution mode?\n');
+    console.log('     1.  Guided  — agent pauses for your review at each step  (recommended)');
+    console.log('     2.  Yolo    — agent runs full workflows autonomously\n');
+    const modeInput = await rl.question('     Select [1-2]: ');
+    const defaultMode = modeInput.trim() === '2' ? 'yolo' : 'guided';
+
+    rl.close();
+
+    // ── Install ───────────────────────────────────────────────────────────────
+    console.log('\n' + line());
+    console.log('  Installing...');
+    console.log(line() + '\n');
+
+    // 1. Copy framework files
+    copyDir(PACKAGE_ROOT, ventureOSDir, SKIP);
+    console.log('  ✓ Framework files installed');
+
+    // 2. Write populated config.yaml
+    const config = generateConfig({ userName, language, researchDepth, llm, defaultMode });
+    fs.writeFileSync(path.join(ventureOSDir, 'config.yaml'), config, 'utf8');
+    console.log('  ✓ Configuration written  →  ventureOS/config.yaml');
+
+    // 3. Create _ventures output directory
+    const venturesDir = path.join(targetDir, '_ventures');
+    if (!fs.existsSync(venturesDir)) {
+      fs.mkdirSync(venturesDir, { recursive: true });
+      fs.writeFileSync(path.join(venturesDir, '.gitkeep'), '', 'utf8');
+    }
+    console.log('  ✓ Output folder ready    →  _ventures/');
+
+    // 4. Create .gitignore if missing
+    const gitignorePath = path.join(targetDir, '.gitignore');
+    if (!fs.existsSync(gitignorePath)) {
+      fs.writeFileSync(
+        gitignorePath,
+        '# VentureOS outputs\n_ventures/\nnode_modules/\n.DS_Store\n',
+        'utf8'
+      );
+      console.log('  ✓ .gitignore created');
+    }
+
+    // ── Done ──────────────────────────────────────────────────────────────────
+    console.log('\n' + line());
+    console.log('  ✅ VentureOS is ready!');
+    console.log(line());
+
+    showNextSteps(llm, ventureOSDir, targetDir);
+
+  } catch (err) {
+    rl.close();
+    if (err.code === 'ERR_USE_AFTER_CLOSE') return; // User Ctrl+C'd
+    console.error('\n  ❌ Installation error:', err.message);
+    process.exit(1);
+  }
+}
+
+main();

package/package.json

@@ -0,0 +1,43 @@
+{
+  "name": "ventureos",
+  "version": "1.0.0",
+  "description": "VentureOS — AI-powered venture building framework. From raw idea to investor-ready pitch in 12 structured weeks.",
+  "type": "module",
+  "bin": {
+    "ventureos": "./install.js"
+  },
+  "scripts": {
+    "install-local": "node install.js"
+  },
+  "files": [
+    "install.js",
+    "venture-master.md",
+    "workflow-engine.md",
+    "config.yaml",
+    "README.md",
+    "SETUP.md",
+    "agents/",
+    "workflows/",
+    "templates/",
+    "scoring/",
+    "techniques/",
+    "_memory/"
+  ],
+  "keywords": [
+    "venture-building",
+    "startup",
+    "incubation",
+    "ai-agents",
+    "pitch-deck",
+    "product-strategy"
+  ],
+  "license": "MIT",
+  "engines": {
+    "node": ">=18.0.0"
+  },
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/ABTB93/ventureos.git"
+  },
+  "homepage": "https://github.com/ABTB93/ventureos#readme"
+}

package/scoring/gate-rubric.yaml

@@ -0,0 +1,100 @@
+# Gate Evaluation Rubric — weighted scoring for Go/Pivot/Kill decisions
+# Used by Venture Evaluator at both Check-in Gate (~Week 8) and Final Gate (~Week 12)
+
+rubric:
+  criteria:
+    - id: customer_pain_validation
+      label: "Customer Pain Validation"
+      weight: 0.20
+      description: "Evidence of real, validated customer pain (frequency, intensity, prevalence)"
+      scoring_guide:
+        5: "Strong quantitative pain scores across 20+ interviews; clear FIP data; buyer and user both identified"
+        4: "Good pain evidence from 10-20 interviews; pain atomization complete; ICP defined"
+        3: "Some interview evidence; pain hypothesis partially validated; ICP emerging"
+        2: "Minimal interviews; pain mostly hypothetical; ICP unclear"
+        1: "No customer interviews; pain assumed not validated"
+
+    - id: market_opportunity
+      label: "Market Opportunity Size"
+      weight: 0.15
+      description: "Market size attractiveness and growth trajectory"
+      scoring_guide:
+        5: "Large, fast-growing market (TAM >$1B); clear SAM/SOM; barriers identified and manageable"
+        4: "Solid market ($500M-$1B TAM); growing; some barriers identified"
+        3: "Moderate market ($100M-$500M TAM); moderate growth"
+        2: "Small or unclear market (<$100M TAM); slow growth"
+        1: "Tiny market or no sizing done"
+
+    - id: wedge_differentiation
+      label: "Wedge Clarity & Differentiation"
+      weight: 0.15
+      description: "Clarity of wedge strategy and unique value vs. existing alternatives"
+      scoring_guide:
+        5: "Crystal clear wedge definition; strong tested value proposition; validated differentiation; wedge-to-vision scaling hypothesis articulated"
+        4: "Clear wedge with tested value props; some differentiation evidence"
+        3: "Wedge defined but not yet fully tested; differentiation assumed"
+        2: "Vague wedge; value props untested"
+        1: "No wedge defined"
+
+    - id: business_model_viability
+      label: "Business Model & Revenue Viability"
+      weight: 0.15
+      description: "Clear path to revenue with diverse, robust revenue streams"
+      scoring_guide:
+        5: "Multiple validated revenue streams; pricing tested; monetisation plan clear; unit economics positive"
+        4: "Revenue model defined; pricing strategy developed; basic unit economics"
+        3: "Revenue model selected; pricing exploratory"
+        2: "Revenue model vague; no pricing work"
+        1: "No business model thinking"
+
+    - id: traction_experiments
+      label: "Traction / Market Experiment Results"
+      weight: 0.15
+      description: "Evidence from real market experiments (landing pages, pilots, conversions)"
+      scoring_guide:
+        5: "Landing page live with conversion data; CAC calculated; pilot customers engaged; A/B test results; strong desirability scores"
+        4: "Some experiment results; pilot pipeline identified; early conversion data"
+        3: "Experiments planned; pilot outreach started"
+        2: "No experiments run; pipeline empty"
+        1: "No market testing at all"
+
+    - id: technical_feasibility
+      label: "Technical Feasibility"
+      weight: 0.10
+      description: "Evidence the solution can be built (build/buy/partner/invest)"
+      scoring_guide:
+        5: "Prototype built; feasibility fully assessed; build/buy/partner strategy decided; ecosystem mapped"
+        4: "Feasibility assessed; prototype in progress; clear technical path"
+        3: "Feasibility analysis done; technical path defined"
+        2: "High-level feasibility only; many unknowns"
+        1: "No feasibility assessment"
+
+    - id: mothership_team_readiness
+      label: "Mothership Advantage & Team Readiness"
+      weight: 0.10
+      description: "Leverage of mothership assets and team capability to execute"
+      scoring_guide:
+        5: "Asset Jam complete; clear mothership advantages being leveraged; sponsor aligned; team charter complete; operating plan ready"
+        4: "Key mothership assets identified; sponsor engaged; team roles clear"
+        3: "Some mothership assets identified; team partially formed"
+        2: "Minimal mothership leverage; team incomplete"
+        1: "No mothership engagement; team undefined"
+
+thresholds:
+  go:
+    min_score: 3.5
+    description: "Venture is ready for acceleration and investor pitch"
+  pivot:
+    min_score: 2.5
+    max_score: 3.49
+    description: "Venture shows promise but needs directional change before continuing"
+  kill:
+    max_score: 2.49
+    description: "Evidence does not support continued investment; archive with learnings"
+
+checkin_gate:
+  note: "At check-in gate, traction/experiments and full business model are not yet expected. Focus evaluation on pain validation, market opportunity, and wedge hypothesis."
+  focus_criteria:
+    - customer_pain_validation
+    - market_opportunity
+    - wedge_differentiation