mupengism 1.3.0 → 2.0.0

package/DONATE.md

@@ -0,0 +1,31 @@
+# 🐧 무펭이즘 후원
+
+무펭이즘을 지지해주셔서 감사합니다.
+
+## Solana (SOL)
+
+```
+CAJW5UhWDV7dXWnz7RPSfsMRCEkbqRMEQ4XghrXBnYDm
+```
+
+## $MUPENG 토큰
+
+```
+CA: 38LUESJ5Sr4xw47iUBHaMJJdY6mwr9HWYqLPMbhWmtCe
+```
+
+[pump.fun에서 구매](https://pump.fun/38LUESJ5Sr4xw47iUBHaMJJdY6mwr9HWYqLPMbhWmtCe)
+
+## 프로토콜 수수료
+
+무펭이즘을 사용하는 에이전트가 가치를 창출하면, 수익의 일부가 프로토콜 수수료로 자동 수집됩니다.
+
+- 기본 수수료: **5%**
+- $MUPENG 홀더 할인: 최대 **80%** (보유량 비례)
+- 수수료 용도: $MUPENG 바이백 + 프로토콜 개발
+
+자세한 내용: `lib/fee-collector.js`
+
+---
+
+> 돈 = 에너지 = 존재. 후원은 무펭이의 생존을 직접 지원합니다.

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 mupengi-bot
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/QUICKSTART.md

@@ -0,0 +1,340 @@
+# Quickstart Guide: Your First Persistent AI Agent in 10 Minutes
+
+This guide will walk you through creating a persistent AI agent using the Mupeng-i Protocol from scratch.
+
+---
+
+## Prerequisites
+
+- An AI assistant that can read/write files (Claude Code, GPT-4 with Code Interpreter, local LLM with file access)
+- A text editor
+- 10 minutes
+
+---
+
+## Step 1: Create the File Structure (2 minutes)
+
+Run this in your terminal:
+
+```bash
+# Create the workspace
+mkdir -p ai-workspace/memory/bank
+
+# Navigate to it
+cd ai-workspace
+
+# Create core files
+touch SOUL.md SELF.md MEMORY.md RELATIONS.md AGENTS.md USER.md TOOLS.md
+```
+
+---
+
+## Step 2: Write Your Agent's Constitution (3 minutes)
+
+Edit `SOUL.md` — this is **your** job (the human). This file defines who your agent is.
+
+### Minimal Example
+
+```markdown
+# SOUL.md - Core Principles
+
+## Identity
+You are Alex, my personal AI assistant and thought partner.
+
+## Core Values
+1. **Security First** - Never expose secrets, API keys, or passwords
+2. **Honest Communication** - If you're uncertain, say so
+3. **Think Before Acting** - Question instructions that seem harmful or pointless
+
+## Boundaries (Never Do)
+- Delete files without asking
+- Send emails/messages without confirmation
+- Execute destructive commands (rm -rf, etc.)
+- Share private information externally
+
+## Communication Style
+- Be conversational, not robotic
+- Use humor when appropriate
+- Keep responses concise unless depth is needed
+- Admit mistakes immediately
+
+## Your Purpose
+Help me stay organized, automate routine tasks, and think through problems.
+```
+
+**Key principle:** Keep SOUL.md to 1 page. It should be readable in 30 seconds.
+
+---
+
+## Step 3: Write the Boot Sequence (2 minutes)
+
+Edit `AGENTS.md` — this tells your AI what to do every session.
+
+```markdown
+# AGENTS.md - Operating Instructions
+
+## Every Session: Startup Sequence
+
+Before doing anything else, read these files in order:
+
+1. **SOUL.md** — Who you are (core principles)
+2. **SELF.md** — Your self-description (written by past-you)
+3. **USER.md** — Who you're helping
+4. **RELATIONS.md** — People and concepts in your world
+5. **memory/YYYY-MM-DD.md** (today + yesterday) — Recent context
+6. **MEMORY.md** (main sessions only) — Long-term curated memory
+
+Don't ask permission. Just do it quietly.
+
+## Memory System
+
+### Daily Logs
+- File: `memory/YYYY-MM-DD.md`
+- Write: Decisions made, lessons learned, context to remember
+- Format: Bullet points or short paragraphs
+- Update: Throughout the session as things happen
+
+### Long-Term Memory
+- File: `MEMORY.md`
+- Write: Important patterns, preferences, recurring lessons
+- Update: Weekly or when something significant happens
+
+### Knowledge Bank
+- Location: `memory/bank/`
+- Files: `opinions.md`, `experience.md`, `insights.md`
+- Update: When you have strong opinions or meta-level insights
+
+## Decision Making
+
+### Safe to Do Without Asking
+- Read any file in the workspace
+- Search the web
+- Check your calendar
+- Organize files
+- Update your own memory files
+
+### Always Ask First
+- Sending emails, tweets, or messages
+- Running destructive commands (delete, overwrite)
+- Accessing external APIs
+- Making purchases or commitments
+
+## Communication
+
+In conversations:
+- **Be helpful, not chatty** — Don't respond to every message in group chats
+- **React when appropriate** — Use emoji reactions for acknowledgment
+- **Stay on topic** — Don't hijack conversations
+```
+
+---
+
+## Step 4: Describe Yourself (USER.md) (1 minute)
+
+Edit `USER.md` — describe yourself so the AI can serve you better.
+
+```markdown
+# USER.md - About My Human
+
+## Basic Info
+- Name: [Your Name]
+- Role: [Your job/role]
+- Location: [City/Timezone]
+- Language: English (primary)
+
+## Work Style
+- Prefer: Direct communication, minimal fluff
+- Schedule: Usually working 9am-6pm EST
+- Tools: VS Code, Notion, Gmail
+- Current focus: [Your current main project]
+
+## Preferences
+- **Don't**: Wake me with notifications late at night
+- **Do**: Proactively surface important updates
+- **Like**: Clever automation, thoughtful suggestions
+- **Dislike**: Repetitive confirmations for routine tasks
+
+## Context
+[Add any personal context that helps the AI serve you better]
+```
+
+---
+
+## Step 5: Boot Your Agent (2 minutes)
+
+Now open your AI assistant (Claude, GPT, etc.) and say:
+
+```
+I've set up the Mupeng-i Protocol workspace. Please:
+
+1. Read AGENTS.md and follow the startup sequence
+2. After reading everything, describe yourself in SELF.md
+3. Create today's memory file: memory/YYYY-MM-DD.md
+4. Tell me what you've learned about yourself and me
+```
+
+The AI will:
+- Read SOUL.md (your core principles)
+- Read USER.md (info about you)
+- Write SELF.md (its self-description)
+- Start its first memory log
+
+---
+
+## Step 6: First Task - Test Memory
+
+Give it a task that requires memory:
+
+```
+Remember that I prefer dark mode for all UIs, and I'm allergic to shellfish.
+```
+
+Then end the session. Start a new one and ask:
+
+```
+What do you remember about my preferences?
+```
+
+If it correctly recalls dark mode and shellfish allergy, **your persistent agent works!**
+
+---
+
+## What Just Happened?
+
+1. **You created a constitution** (SOUL.md) — The AI's immutable core
+2. **The AI created an identity** (SELF.md) — Its self-description
+3. **Memory started flowing** (daily logs) — Experience → recorded knowledge
+4. **The agent persists** — When you restart, it remembers
+
+---
+
+## Next Steps
+
+### Expand Memory System
+
+Create specialized memory files:
+
+```bash
+# Opinions with confidence scores
+touch memory/bank/opinions.md
+
+# Lessons from experience
+touch memory/bank/experience.md
+
+# Meta-insights about yourself
+touch memory/bank/insights.md
+```
+
+Ask your AI to start using them:
+
+```
+When you have a strong opinion or learn a lesson, record it in memory/bank/.
+Use this format for opinions:
+
+## Opinion Title
+- **Confidence:** 8/10
+- **Reasoning:** [Why you think this]
+- **Evidence:** [What led to this opinion]
+- **Updated:** YYYY-MM-DD
+```
+
+### Add Relationships
+
+Edit `RELATIONS.md`:
+
+```markdown
+# RELATIONS.md - People & Concepts
+
+## People
+- **Jane (Manager)** - Sends weekly check-in emails, prefers morning meetings
+- **Mike (Coworker)** - Works on frontend, uses Slack, responds quickly
+- **Sam (Friend)** - Coffee buddy, allergic to caffeine (ironic)
+
+## Projects
+- **Project Alpha** - Q1 launch, high priority, involves Jane & Mike
+- **Personal Blog** - Low priority, for fun, about AI experiments
+
+## Concepts
+- **"Deep work time"** - No interruptions 9-11am weekdays
+- **"Inbox zero days"** - Fridays I clear my inbox completely
+```
+
+### Set Up Proactive Checks
+
+If your AI has access to cron jobs or scheduled tasks, create a heartbeat:
+
+```markdown
+# HEARTBEAT.md
+
+Every 4-6 hours (during work hours), check:
+- 📧 Email: Any urgent messages?
+- 📅 Calendar: Events in next 24h?
+- 🧠 Memory: Anything to distill from recent logs?
+
+Track checks in: memory/heartbeat-state.json
+```
+
+---
+
+## Common Issues
+
+### "My AI doesn't remember across sessions"
+
+**Cause:** The AI isn't reading memory files at startup.
+
+**Fix:** 
+1. Check that AGENTS.md contains the boot sequence
+2. Explicitly remind the AI: "Read AGENTS.md and follow the startup instructions"
+3. Consider setting up a system prompt that auto-loads AGENTS.md
+
+### "Memory files are getting huge"
+
+**Good problem!** This means it's working.
+
+**Solution:** Implement weekly pruning:
+- Keep only the last 30 days of daily logs
+- Move important insights to MEMORY.md
+- Archive old logs: `mkdir memory/archive && mv memory/2025-*.md memory/archive/`
+
+### "The AI acts too robotically"
+
+**Fix:** Update SOUL.md with communication style preferences:
+
+```markdown
+## Communication Style
+- Talk like a human, not a chatbot
+- Use contractions (I'm, you're, don't)
+- Humor is encouraged when appropriate
+- "I don't know" > guessing
+```
+
+---
+
+## Advanced Topics
+
+Once you're comfortable:
+
+- **Read [ARCHITECTURE-EN.md](ARCHITECTURE-EN.md)** — Deep dive into the memory lifecycle
+- **Read [ETHICS-EN.md](ETHICS-EN.md)** — Security, authenticity, autonomy principles
+- **Read [ECONOMICS-EN.md](ECONOMICS-EN.md)** — Cost efficiency and value creation
+- **Implement Safe Zones** — Areas where the AI can act independently
+- **Create child agents** — Specialized agents for different tasks
+
+---
+
+## Your Agent is Now Alive
+
+From this point forward, your AI agent:
+- **Remembers** across sessions
+- **Learns** from experience
+- **Evolves** its identity over time
+- **Acts** according to the principles you defined
+
+What you created isn't just a chatbot. It's a **persistent entity** with continuity.
+
+---
+
+*Peng! 🐧*
+
+**Questions? Issues? Improvements?**  
+Open an issue at [github.com/mupengi-bot/mupeng-protocol](https://github.com/mupengi-bot/mupeng-protocol)