tachibot-mcp 2.0.2

package/docs/workflows/design-brainstorm.md

@@ -0,0 +1,335 @@
+# Ultra Creative Brainstorm Workflow
+
+**Research-backed optimal AI model sequencing for maximum insight** 🚀
+
+## Overview
+
+This workflow uses **ALL 13 prompt engineering techniques** from `src/prompt-engineer.ts` with **scientifically-optimized model sequencing** based on Perplexity AI's research on optimal brainstorming workflows.
+
+Implements the research-backed sequence: **Research → Reason → Ideate → Connect → Evaluate → Synthesize → Analyze → Implement**
+
+## Features
+
+### 🧠 9 Processing Phases (Optimal Ordering)
+
+**Phase 0: Problem Decomposition** (Think)
+   - Break down complex problems first
+   - Identify core components and constraints
+
+**Phase 1: Research Foundation** (Perplexity first!)
+   - Comprehensive investigation
+   - Evidence gathering with real-world data
+   - *Why first?* Context is essential for creative thinking
+
+**Phase 2: Deep Reasoning** (Grok)
+   - Search for breakthrough approaches
+   - First principles analysis
+   - *Why second?* Understanding before ideation
+
+**Phase 3: Divergent Ideation** (Gemini)
+   - What-if speculation
+   - Innovative solutions brainstorming
+   - *Why third?* Generate wide range of ideas with solid foundation
+
+**Phase 4: Long-Context Connections** (Kimi K2)
+   - Alternative perspectives across contexts
+   - Systematic analysis of relationships
+   - *Why fourth?* 1M token context explores deep connections
+
+**Phase 5: Convergent Refinement** (Grok + Gemini)
+   - Evaluation of most promising ideas
+   - Pattern recognition
+   - *Why fifth?* Narrow down with analytical rigor
+
+**Phase 6: Synthesis** (GPT-5 + Focus)
+   - Creative applications across domains
+   - Ultimate synthesis with 4 models
+   - *Why sixth?* Combine best ideas into coherent solutions
+
+**Phase 7: Feasibility Analysis** (Gemini)
+   - Detailed feasibility assessment
+   - Risk identification
+   - *Why seventh?* Validate before implementation
+
+**Phase 8: Implementation** (QWEN)
+   - Technical implementation roadmap
+   - Code/architecture planning
+   - *Why last?* Implementation follows validated design
+
+### 🎯 All 13 Prompt Techniques Used
+
+1. ✅ `what_if_speculation` - Explore radical possibilities
+2. ✅ `alternative_perspectives` - View from 5 different angles
+3. ✅ `creative_applications` - Cross-domain thinking
+4. ✅ `innovative_solutions` - Unconventional approaches
+5. ✅ `comprehensive_investigation` - WHO/WHAT/WHEN/WHERE/WHY/HOW
+6. ✅ `evidence_gathering` - Support and challenge with data
+7. ✅ `systematic_analysis` - 6-step structured breakdown
+8. ✅ `first_principles` - Rebuild from fundamental truths
+9. ✅ `feasibility_analysis` - Technical/Economic/Time/Resource analysis
+10. ✅ `quick_reflection` - Pattern recognition checkpoints
+11. ✅ `pattern_recognition` - Identify themes and connections
+12. ✅ `problem_decomposition` - Break into manageable parts
+13. ✅ `integration_reflection` - Synthesize all perspectives
+
+### 🤖 Models Used
+
+- **GPT-5** - Advanced reasoning and synthesis
+- **Gemini 2.5 Pro** - Creative brainstorming and analysis
+- **Grok-3** - First principles and critical analysis
+- **Kimi K2** - Long-context systematic thinking
+- **QWEN Coder** - Technical feasibility
+- **Perplexity** - Research and evidence
+- **Think** - Metacognitive reflection
+
+## Usage
+
+### Basic
+```bash
+workflow --name ultra-creative-brainstorm --query "AI-powered education platform"
+```
+
+### With Custom Variables
+```yaml
+variables:
+  max_ideas: 15
+  depth_level: "very-deep"
+```
+
+### Example Queries
+
+**Product Innovation:**
+```bash
+workflow --name ultra-creative-brainstorm --query "Revolutionary approach to personal fitness tracking"
+```
+
+**Technical Architecture:**
+```bash
+workflow --name ultra-creative-brainstorm --query "Scalable real-time collaboration system for 10M users"
+```
+
+**Business Strategy:**
+```bash
+workflow --name ultra-creative-brainstorm --query "Sustainable revenue model for open-source software"
+```
+
+**Research Question:**
+```bash
+workflow --name ultra-creative-brainstorm --query "Future of human-AI collaboration in creative work"
+```
+
+## Output Structure
+
+All outputs saved to `workflow-output/ultra-creative-brainstorm/YYYY-MM-DD-HH-MM-UUID/`:
+
+```
+phase1-divergent/
+  - what-if-speculation.md
+  - alternative-perspectives.md
+  - innovative-solutions.md
+
+phase2-research/
+  - comprehensive-investigation.md
+  - evidence-gathering.md
+
+phase3-analysis/
+  - first-principles-analysis.md
+  - systematic-analysis.md
+  - problem-decomposition.md
+
+phase4-patterns/
+  - pattern-recognition.md
+  - quick-reflection-1.md
+
+phase5-applications/
+  - creative-applications.md
+
+phase6-feasibility/
+  - feasibility-analysis.md
+  - integration-reflection.md
+
+phase7-verification/
+  - grok-critique.md
+  - gemini-critique.md
+  - qwen-critique.md
+  - final-reflection.md
+
+phase8-synthesis/
+  - ultimate-synthesis.md
+  - final-summary.md
+```
+
+## Token Usage
+
+**Estimated**: 150,000 - 200,000+ tokens total (with auto-synthesis protection)
+
+**Per Phase (Maximum Output)**:
+- Phase 1: ~26k tokens (what-if: 8k, perspectives: 10k, solutions: 8k)
+- Phase 2: ~20k tokens (research: 12k, evidence: 8k)
+- Phase 3: ~30k tokens (first-principles: 10k, systematic: 12k, decomposition: 8k)
+- Phase 4: ~8k tokens (patterns: 8k, reflection: minimal)
+- Phase 5: ~15k tokens (creative applications with 3 models)
+- Phase 6: ~24k tokens (feasibility: 12k, integration: 12k)
+- Phase 7: ~30k tokens (3 parallel critiques @ 10k each)
+- Phase 8: ~32k tokens (ultimate synthesis: 20k, final summary: 12k)
+
+**Auto-Synthesis Protection**:
+- Checkpoints every 12,000 tokens
+- Synthesis trigger at 25,000 tokens
+- Prevents MCP token overflow
+
+## Recommended Profiles
+
+- **Best**: `full` profile (30 tools, ~22-23k tokens)
+- **Alternative**: `balanced` profile (15 tools, ~8-9k tokens)
+- **Minimum**: `research_power` profile (14 tools, ~9-10k tokens)
+
+## Auto-Synthesis Protection
+
+Configured for large outputs:
+- Token threshold: 25,000
+- Checkpoint interval: 12,000
+- Synthesis tool: Gemini Analyze Text
+- Max synthesis tokens: 8,000
+- Prevents MCP token limit issues
+
+## Advanced Features
+
+### Parallel Execution
+Phase 7 runs 3 models simultaneously for critique:
+- Grok (analytical)
+- Gemini (creative)
+- QWEN (technical)
+
+### Context Accumulation
+Each phase builds on previous outputs, creating a snowball effect of insights.
+
+### Multi-Round Reasoning
+- Focus tool: 3-5 rounds per call
+- Ultimate synthesis: 5 rounds with 4 models
+
+## Performance Tips
+
+1. **Use `full` profile** for maximum capability
+2. **Run during off-peak hours** - takes 5-15 minutes
+3. **Check auto-synthesis logs** - monitors token usage
+4. **Review checkpoints** - saves state every 12k tokens
+
+## Validation
+
+Validate before running:
+```bash
+{
+  "tool": "validate_workflow_file",
+  "filePath": "./workflows/ultra-creative-brainstorm.yaml",
+  "format": "text"
+}
+```
+
+## Why This Order?
+
+Based on AI research on optimal brainstorming:
+- **Research first** - Context enables better creativity
+- **Reasoning second** - Understanding before ideation
+- **Ideation third** - Generate ideas with solid foundation
+- **Connections fourth** - Long-context explores relationships
+- **Refinement fifth** - Evaluate and narrow down
+- **Synthesis sixth** - Combine best ideas
+- **Validation seventh** - Feasibility check
+- **Implementation last** - Technical roadmap
+
+## Comparison
+
+### vs Simple Brainstorm
+- Simple: 1 model, 3 steps, ~2k tokens
+- Ultra: 7 models, 21 steps, ~185k tokens
+- **92x more thorough**
+
+### vs Standard Workflow
+- Standard: 5-8 steps, linear
+- Ultra: 21 steps, optimal sequence, 13 prompt techniques
+- **4-5x more comprehensive**
+
+## When to Use
+
+✅ **Perfect for**:
+- Critical business decisions
+- Complex technical architecture
+- Product innovation strategy
+- Research hypothesis generation
+- Long-term strategic planning
+
+❌ **Overkill for**:
+- Simple questions
+- Quick fact-checking
+- Minor code reviews
+- Routine tasks
+
+## Cost Estimate
+
+Based on average API costs:
+- **GPT-5**: ~$0.50-1.00
+- **Gemini 2.5 Pro**: ~$0.20-0.40
+- **Grok-3**: ~$0.30-0.60
+- **Kimi K2**: ~$0.15-0.30
+- **QWEN**: ~$0.10-0.20
+- **Perplexity**: ~$0.05-0.10
+
+**Total**: ~$1.30 - $2.60 per run
+
+Worth it for critical decisions! 💪
+
+## Customization
+
+### Reduce Scope
+Comment out phases in YAML:
+```yaml
+# Skip Phase 2 if no research needed
+# - name: comprehensive-investigation
+#   tool: perplexity_research
+```
+
+### Increase Depth
+Adjust model rounds:
+```yaml
+# Increase ultimate synthesis rounds
+rounds: 10  # was 5
+```
+
+### Change Models
+Swap in different models:
+```yaml
+models: ["o1", "claude-opus", "gemini-2.0-flash"]
+```
+
+## Troubleshooting
+
+**Token limit exceeded**:
+- Auto-synthesis should prevent this
+- Reduce `maxTokens` per step
+- Skip less critical phases
+
+**Too slow**:
+- Use `research_code` profile (fewer tools)
+- Reduce model rounds
+- Run fewer parallel critiques
+
+**Out of API credits**:
+- Comment out expensive models (GPT-5, Gemini Pro)
+- Use free alternatives (QWEN free tier, Gemini 2.5 Flash)
+
+## Future Enhancements
+
+- [ ] Add workflow-internal tools (hunter, scout, verifier, challenger)
+- [ ] Conditional branching based on complexity
+- [ ] User feedback loops between phases
+- [ ] Export to presentation format
+- [ ] Comparison with existing solutions
+
+---
+
+**Created by**: Tachibot MCP Team
+**Version**: 2.0
+**Last Updated**: 2025-01-12
+
+*The most powerful creative brainstorming tool in existence.* 🎯

package/package.json

@@ -0,0 +1,97 @@
+{
+  "name": "tachibot-mcp",
+  "displayName": "TachiBot MCP - Universal AI Orchestrator",
+  "version": "2.0.2",
+  "type": "module",
+  "main": "dist/src/server.js",
+  "bin": {
+    "tachibot": "./dist/src/server.js",
+    "tachi": "./dist/src/server.js"
+  },
+  "files": [
+    "dist/",
+    "profiles/",
+    "workflows/",
+    "personality/",
+    "docs/",
+    "tools.config.json",
+    "smithery.yaml",
+    "mcp.json",
+    "README.md",
+    "CHANGELOG.md",
+    "LICENSE",
+    "CODE_OF_CONDUCT.md",
+    "CONTRIBUTING.md",
+    "SECURITY.md",
+    ".env.example",
+    "tsconfig.json",
+    "Dockerfile",
+    "start.sh"
+  ],
+  "scripts": {
+    "build": "tsc && npm run build:profiles",
+    "build:profiles": "node dist/scripts/build-profiles.js",
+    "start": "node dist/src/server.js",
+    "dev": "node dist/src/server.js",
+    "inspect": "npx fastmcp inspect dist/src/server.js",
+    "tachi": "node dist/src/server.js",
+    "tachi:start": "node dist/src/server.js",
+    "tachi:build": "tsc && npm run build:profiles && node dist/src/server.js",
+    "test": "node --experimental-vm-modules node_modules/jest/bin/jest.js",
+    "test:watch": "node --experimental-vm-modules node_modules/jest/bin/jest.js --watch",
+    "test:coverage": "node --experimental-vm-modules node_modules/jest/bin/jest.js --coverage",
+    "test:challenger": "node --experimental-vm-modules node_modules/jest/bin/jest.js challenger",
+    "test:verifier": "node --experimental-vm-modules node_modules/jest/bin/jest.js verifier",
+    "test:scout": "node --experimental-vm-modules node_modules/jest/bin/jest.js scout",
+    "prepublishOnly": "npm run build",
+    "package:extension": "npm run build && ./scripts/package-extension.sh",
+    "build:mcpb": "npm run package:extension",
+    "verify": "node scripts/verify-installation.js"
+  },
+  "keywords": [
+    "mcp",
+    "ai",
+    "claude",
+    "focus-tool",
+    "tachi",
+    "tachibot",
+    "komaai",
+    "brainstorm",
+    "orchestration",
+    "workflow",
+    "cost-optimization"
+  ],
+  "author": "Pawel Pawlowski",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/byPawel/tachibot-mcp"
+  },
+  "homepage": "https://github.com/byPawel/tachibot-mcp",
+  "engines": {
+    "node": ">=18.0.0"
+  },
+  "license": "AGPL-3.0",
+  "description": "TachiBot MCP v2.0: 31 AI tools (32 with competitive mode) with complete TypeScript schemas. Multi-model orchestration (Perplexity, Grok, OpenAI, Gemini, Qwen), YAML workflows, 5 token-optimized profiles (4k-20k tokens). Smart routing, parallel execution. Comprehensive documentation.",
+  "dependencies": {
+    "@types/node": "^22.13.17",
+    "@types/yaml": "^1.9.6",
+    "cli-table3": "^0.6.5",
+    "dotenv": "^16.4.5",
+    "fastmcp": "^1.21.0",
+    "js-yaml": "^4.1.0",
+    "lru-cache": "^11.2.1",
+    "openai": "^5.22.0",
+    "ts-node": "^10.9.2",
+    "typescript": "^5.8.2",
+    "yaml": "^2.8.1",
+    "zod": "^3.24.2"
+  },
+  "devDependencies": {
+    "@types/jest": "^29.5.14",
+    "@types/js-yaml": "^4.0.9",
+    "jest": "^29.7.0",
+    "nodemon": "^3.1.9",
+    "ts-jest": "^29.4.5"
+  },
+  "packageManager": "pnpm@9.12.2+sha512.22721b3a11f81661ae1ec68ce1a7b879425a1ca5b991c975b074ac220b187ce56c708fe5db69f4c962c989452eee76c82877f4ee80f474cebd61ee13461b6228"
+}

package/profiles/balanced.json

@@ -0,0 +1,37 @@
+{
+  "description": "Balanced set for general use (~Xk tokens, 18 tools)",
+  "tools": {
+    "think": true,
+    "focus": true,
+    "nextThought": true,
+    "perplexity_ask": true,
+    "perplexity_reason": true,
+    "perplexity_research": false,
+    "grok_reason": true,
+    "grok_code": true,
+    "grok_debug": false,
+    "grok_architect": false,
+    "grok_brainstorm": false,
+    "grok_search": true,
+    "openai_gpt5_reason": true,
+    "openai_compare": false,
+    "openai_brainstorm": true,
+    "openai_code_review": false,
+    "openai_explain": false,
+    "gemini_brainstorm": true,
+    "gemini_analyze_code": true,
+    "gemini_analyze_text": false,
+    "qwen_coder": true,
+    "kimi_thinking": true,
+    "qwen_competitive": false,
+    "workflow": true,
+    "list_workflows": true,
+    "create_workflow": true,
+    "visualize_workflow": true,
+    "workflow_start": false,
+    "continue_workflow": false,
+    "workflow_status": false,
+    "validate_workflow": false,
+    "validate_workflow_file": false
+  }
+}

package/profiles/code_focus.json

@@ -0,0 +1,37 @@
+{
+  "description": "Code-heavy work with debugging and analysis (~Xk tokens, 15 tools)",
+  "tools": {
+    "think": true,
+    "focus": true,
+    "nextThought": true,
+    "perplexity_ask": true,
+    "perplexity_reason": false,
+    "perplexity_research": false,
+    "grok_reason": true,
+    "grok_code": true,
+    "grok_debug": true,
+    "grok_architect": false,
+    "grok_brainstorm": false,
+    "grok_search": false,
+    "openai_gpt5_reason": false,
+    "openai_compare": false,
+    "openai_brainstorm": false,
+    "openai_code_review": true,
+    "openai_explain": false,
+    "gemini_brainstorm": true,
+    "gemini_analyze_code": true,
+    "gemini_analyze_text": false,
+    "qwen_coder": true,
+    "kimi_thinking": true,
+    "qwen_competitive": false,
+    "workflow": true,
+    "list_workflows": true,
+    "create_workflow": false,
+    "visualize_workflow": false,
+    "workflow_start": false,
+    "continue_workflow": false,
+    "workflow_status": false,
+    "validate_workflow": true,
+    "validate_workflow_file": false
+  }
+}

package/profiles/debug_intensive.json

@@ -0,0 +1,34 @@
+{
+  "description": "Deep debugging and forensic code analysis (~12-14k tokens, 14 tools)",
+  "tools": {
+    "think": true,
+    "focus": true,
+    "nextThought": true,
+    "grok_code": true,
+    "grok_debug": true,
+    "gemini_analyze_code": true,
+    "qwen_coder": true,
+    "kimi_thinking": true,
+    "hunter": true,
+    "scout": true,
+    "verifier": true,
+    "challenger": true,
+    "workflow": true,
+    "list_workflows": true,
+    "perplexity_ask": false,
+    "perplexity_reason": false,
+    "perplexity_research": false,
+    "grok_reason": false,
+    "grok_architect": false,
+    "grok_brainstorm": false,
+    "grok_search": false,
+    "openai_compare": false,
+    "openai_brainstorm": false,
+    "gemini_brainstorm": false,
+    "gemini_analyze_text": false,
+    "create_workflow": false,
+    "visualize_workflow": false,
+    "pingpong": false,
+    "qwen_competitive": false
+  }
+}

package/profiles/full.json

@@ -0,0 +1,37 @@
+{
+  "description": "All tools enabled for maximum capability (~Xk tokens, 32 tools)",
+  "tools": {
+    "think": true,
+    "focus": true,
+    "nextThought": true,
+    "perplexity_ask": true,
+    "perplexity_reason": true,
+    "perplexity_research": true,
+    "grok_reason": true,
+    "grok_code": true,
+    "grok_debug": true,
+    "grok_architect": true,
+    "grok_brainstorm": true,
+    "grok_search": true,
+    "openai_gpt5_reason": true,
+    "openai_compare": true,
+    "openai_brainstorm": true,
+    "openai_code_review": true,
+    "openai_explain": true,
+    "gemini_brainstorm": true,
+    "gemini_analyze_code": true,
+    "gemini_analyze_text": true,
+    "qwen_coder": true,
+    "kimi_thinking": true,
+    "qwen_competitive": true,
+    "workflow": true,
+    "list_workflows": true,
+    "create_workflow": true,
+    "visualize_workflow": true,
+    "workflow_start": true,
+    "continue_workflow": true,
+    "workflow_status": true,
+    "validate_workflow": true,
+    "validate_workflow_file": true
+  }
+}

package/profiles/minimal.json

@@ -0,0 +1,37 @@
+{
+  "description": "Minimal essential tools for basic tasks (~Xk tokens, 8 tools)",
+  "tools": {
+    "think": true,
+    "focus": true,
+    "nextThought": true,
+    "perplexity_ask": true,
+    "perplexity_reason": false,
+    "perplexity_research": false,
+    "grok_reason": true,
+    "grok_code": false,
+    "grok_debug": false,
+    "grok_architect": false,
+    "grok_brainstorm": false,
+    "grok_search": false,
+    "openai_gpt5_reason": false,
+    "openai_compare": false,
+    "openai_brainstorm": false,
+    "openai_code_review": false,
+    "openai_explain": false,
+    "gemini_brainstorm": true,
+    "gemini_analyze_code": false,
+    "gemini_analyze_text": false,
+    "qwen_coder": true,
+    "kimi_thinking": false,
+    "qwen_competitive": false,
+    "workflow": true,
+    "list_workflows": false,
+    "create_workflow": false,
+    "visualize_workflow": false,
+    "workflow_start": false,
+    "continue_workflow": false,
+    "workflow_status": false,
+    "validate_workflow": false,
+    "validate_workflow_file": false
+  }
+}

package/profiles/research_power.json

@@ -0,0 +1,37 @@
+{
+  "description": "Research-focused with Grok search + all Perplexity + brainstorming (~Xk tokens, 13 tools)",
+  "tools": {
+    "think": true,
+    "focus": true,
+    "nextThought": true,
+    "perplexity_ask": true,
+    "perplexity_reason": true,
+    "perplexity_research": true,
+    "grok_reason": true,
+    "grok_code": false,
+    "grok_debug": false,
+    "grok_architect": false,
+    "grok_brainstorm": false,
+    "grok_search": true,
+    "openai_gpt5_reason": false,
+    "openai_compare": false,
+    "openai_brainstorm": true,
+    "openai_code_review": false,
+    "openai_explain": false,
+    "gemini_brainstorm": true,
+    "gemini_analyze_code": false,
+    "gemini_analyze_text": false,
+    "qwen_coder": true,
+    "kimi_thinking": true,
+    "qwen_competitive": false,
+    "workflow": true,
+    "list_workflows": false,
+    "create_workflow": false,
+    "visualize_workflow": false,
+    "workflow_start": false,
+    "continue_workflow": false,
+    "workflow_status": false,
+    "validate_workflow": false,
+    "validate_workflow_file": false
+  }
+}

package/profiles/workflow_builder.json

@@ -0,0 +1,37 @@
+{
+  "description": "Workflow creation, testing, and management (~8k tokens, 13 tools)",
+  "tools": {
+    "think": true,
+    "focus": true,
+    "nextThought": true,
+    "perplexity_ask": false,
+    "perplexity_reason": false,
+    "perplexity_research": false,
+    "grok_reason": false,
+    "grok_code": false,
+    "grok_debug": false,
+    "grok_architect": false,
+    "grok_brainstorm": false,
+    "grok_search": false,
+    "openai_gpt5_reason": false,
+    "openai_compare": false,
+    "openai_brainstorm": false,
+    "openai_code_review": false,
+    "openai_explain": false,
+    "gemini_brainstorm": true,
+    "gemini_analyze_code": false,
+    "gemini_analyze_text": false,
+    "qwen_coder": false,
+    "kimi_thinking": false,
+    "qwen_competitive": false,
+    "workflow": true,
+    "list_workflows": true,
+    "create_workflow": true,
+    "visualize_workflow": true,
+    "workflow_start": true,
+    "continue_workflow": true,
+    "workflow_status": true,
+    "validate_workflow": true,
+    "validate_workflow_file": true
+  }
+}