shashank2577/hesoyam-for-claude-code

██╗ ██╗███████╗███████╗ ██████╗ ██╗ ██╗ █████╗ ███╗ ███╗ ██║ ██║██╔════╝██╔════╝██╔═══██╗╚██╗ ██╔╝██╔══██╗████╗ ████║ ███████║█████╗ ███████╗██║ ██║ ╚████╔╝ ███████║██╔████╔██║ ██╔══██║██╔══╝ ╚════██║██║ ██║ ╚██╔╝ ██╔══██║██║╚██╔╝██║ ██║ ██║███████╗███████║╚██████╔╝ ██║ ██║ ██║██║ ╚═╝ ██║ ╚═╝ ╚═╝╚══════╝╚══════╝ ╚═════╝ ╚═╝ ╚═╝ ╚═╝╚═╝ ╚═╝ for Claude Code

🎮 Health. Armor. $250K. The cheat code for Claude Code.

In GTA San Andreas, typing HESOYAM gave you full health, full armor, and $250,000 — instantly. No grinding. No side quests. Just everything you need to dominate.

This repo is HESOYAM for Claude Code. One install, and you get orchestration, configuration, memory, knowledge, and ecosystem discovery — the complete cheat code for AI-powered development.

Health (Memory) · Armor (Security) · $250K (Productivity) · Cheat Activated

The Cheats · Quick Start · The $250K · The Armor · The Health Bar · The Safe House · The Map · Contributing

🛤️ Don’t need all five pillars? Read Pick Your Path to find the right setup for you — from minimal (Claude Code only, 15 min) to full stack (everything, 45 min). Obsidian is optional. Every path works without it.

---

🎮 What Is This?

Remember the feeling of typing a cheat code and watching everything change? That’s this repo.

Most Claude Code repos solve one problem. HESOYAM solves all five — like entering a cheat code that gives you max health, infinite ammo, AND a tank at the same time.

You shouldn’t need six browser tabs, four GitHub repos, and a prayer to get Claude Code working the way your brain thinks. HESOYAM combines the best open-source projects in the Claude Code ecosystem — each one legendary in its own right — into a unified reference architecture with opinionated guides, install scripts, and interoperability patterns that make them work together.

Think of it like the GTA cheat sheet you kept bookmarked:

This isn’t a fork. It’s a cheat code.

Important: HESOYAM is a bootstrap kit and reference architecture — it coordinates installation and configuration of best-in-class upstream projects. The agents, skills, and orchestration features come from those upstream projects (oh-my-claudecode, everything-claude-code, claude-mem, etc.). HESOYAM itself provides: the unified installer, 4 integration skills, 6 playbooks, 8 stack configs, 11 guides, an Obsidian vault template, and the glue that makes everything work together.

---

🏙️ How It All Fits Together

flowchart TB
    subgraph "🎮 HESOYAM — The Cheat Code for Claude Code"
        direction TB

        DEV["🧑‍💻 You (the developer)"]

        subgraph ORCH["⚔️ ORCHESTRATION — The Weapons"]
            OMC["oh-my-claudecode"]
            MODES["Autopilot | Ralph | Team Mode | Ultrapilot"]
            OMC --> MODES
        end

        subgraph CONFIG["🛡️ CONFIG & SECURITY — The Armor"]
            ECC["everything-claude-code"]
            AGENTS["28 Agents + 125 Skills"]
            HOOKS["Security Hooks + AgentShield"]
            ECC --> AGENTS
            ECC --> HOOKS
        end

        subgraph MEM["🧠 MEMORY — The Health Bar"]
            CM["claude-mem (auto-capture)"]
            DB["SQLite: ~/.claude-mem/"]
            CM --> DB
        end

        subgraph KNOW["📚 KNOWLEDGE — The Safe House"]
            OBS["Obsidian Vault (PARA)"]
            GIT["Git Repo (auto-push)"]
            PHONE["📱 Phone Access"]
            OBS --> GIT --> PHONE
        end

        subgraph DISC["🔭 DISCOVERY — The Map"]
            AWE["awesome-claude-code"]
            ECO["Ecosystem Catalogs"]
            AWE --> ECO
        end

        DEV --> ORCH
        DEV --> CONFIG
        ORCH --> MEM
        CONFIG --> MEM
        MEM -->|"/promote-to-vault"| KNOW
        DISC -->|"monthly /ecosystem-check"| CONFIG
    end

    style ORCH fill:#FF6B35,color:#fff
    style CONFIG fill:#00D4AA,color:#fff
    style MEM fill:#FFD700,color:#000
    style KNOW fill:#00B8D4,color:#fff
    style DISC fill:#E040FB,color:#fff

---

🚲 ➡️ ✈️ Day 1 vs. Day 50 — What Changes

🚲 Day 1: CJ on a bicycle — You just typed HESOYAM

Your Claude Code setup:
├── .claude/
│   └── skills/                    ← 4 HESOYAM skills installed
│       ├── promote-to-vault/SKILL.md
│       ├── daily-standup/SKILL.md
│       ├── research-sprint/SKILL.md
│       └── ecosystem-check/SKILL.md
├── ~/Documents/obsidian-vault/    ← Empty PARA template
│   ├── 00_Inbox/                  (empty)
│   ├── 01_Projects/               (empty)
│   ├── 02_Areas/                  (empty)
│   ├── 03_Resources/              (empty)
│   ├── 04_Archive/                (empty)
│   └── 06_Templates/             ← 4 templates ready
└── ~/.claude-mem/
    └── claude-mem.db              ← First session captured

Total knowledge: ~0
Claude knows about you: nothing
Vault commits: 1 (initial template)

How it feels: 🚲 Like starting a new GTA save. Clean map. No territory. But you’ve got the cheat codes entered.

✈️ Day 50: CJ in a Hydra jet — Two months of HESOYAM

Your Claude Code setup:
├── .claude/
│   └── skills/                    ← 4 skills + any custom ones you added
├── ~/Documents/obsidian-vault/    ← A living knowledge base
│   ├── 00_Inbox/
│   │   └── random-thought-2026-03-28.md
│   ├── 01_Projects/
│   │   ├── acme-api/
│   │   │   ├── decisions/
│   │   │   │   ├── adr-001-chose-postgresql.md
│   │   │   │   ├── adr-002-jwt-over-sessions.md
│   │   │   │   └── adr-003-event-sourcing-for-billing.md
│   │   │   ├── meetings/
│   │   │   │   ├── 2026-02-03-kickoff.md
│   │   │   │   ├── 2026-02-17-sprint-review.md
│   │   │   │   └── 2026-03-01-architecture-review.md
│   │   │   └── dailies/
│   │   │       ├── 2026-03-25.md
│   │   │       ├── 2026-03-26.md
│   │   │       └── 2026-03-27.md
│   │   ├── mobile-app/
│   │   │   ├── decisions/
│   │   │   │   └── adr-001-react-native-over-flutter.md
│   │   │   └── spike-offline-sync.md
│   │   └── internal-dashboard/
│   │       └── decisions/
│   │           └── adr-001-chose-nextjs.md
│   ├── 02_Areas/
│   │   ├── architecture/
│   │   │   ├── microservices-patterns.md
│   │   │   └── api-versioning-strategy.md
│   │   └── team-management/
│   │       └── onboarding-checklist.md
│   ├── 03_Resources/
│   │   ├── debugging-journal/
│   │   │   ├── 2026-02-12-payment-webhook-race.md
│   │   │   ├── 2026-02-28-n+1-query-in-dashboard.md
│   │   │   └── 2026-03-15-jwt-expiry-edge-case.md
│   │   ├── patterns/
│   │   │   ├── retry-with-exponential-backoff.md
│   │   │   ├── repository-pattern-sqlalchemy.md
│   │   │   └── react-error-boundary-pattern.md
│   │   └── tech-decisions/
│   │       ├── evaluation-queue-systems.md
│   │       └── evaluation-search-engines.md
│   ├── 04_Archive/
│   │   └── old-auth-system/           ← Completed migration
│   └── 06_Templates/
└── ~/.claude-mem/
    └── claude-mem.db              ← 50 days of sessions

Total knowledge: 47 notes, 8 ADRs, 3 debug journals, 3 patterns
Claude knows about you: your stack, your decisions, your bugs, your patterns
Vault commits: 200+ (auto-synced)
On your phone: everything, always

How it feels: ✈️ Like having a 5-star wanted level cleared. You’ve mapped the city. You own territory. Claude Code starts every session already knowing your codebase, your decisions, and your debugging history. New team members clone the vault and skip 2 months of context-building.

📸 See It In Action

Real screenshots from a production HESOYAM setup — not mockups.

The cheat code in action: /promote-to-vault classifying and saving knowledge

The Compound Effect

graph LR
    subgraph "Week 1"
        W1["😐 Manual contextevery session"]
    end
    subgraph "Week 4"
        W4["🙂 claude-mem handlessession continuity"]
    end
    subgraph "Week 8"
        W8["😎 Vault has ADRs,patterns, debug logs"]
    end
    subgraph "Month 3+"
        M3["🤯 Claude knows yourcodebase better thanmost humans on the team"]
    end

    W1 --> W4 --> W8 --> M3

    style W1 fill:#ff4444,color:#fff
    style W4 fill:#ffaa00,color:#000
    style W8 fill:#44bb44,color:#fff
    style M3 fill:#00D4AA,color:#fff

---

🎰 The Five Cheats

---

⚡ Quick Start

Option A: Let Claude Do It (No Commands Required)

Don’t want to run scripts? Open Claude Code in this folder and paste this:

I want to set up HESOYAM in this directory. Ask me what I need, recommend a setup, explain it to me plainly, then do it.

Claude will ask you a few questions, recommend the right profile, explain what it’s doing, and handle everything. No terminal knowledge needed.

→ Full conversational setup guide

---

Option B: Guided Install (Script)

Run the installer — it will guide you to the right setup:

git clone https://github.com/Shashank2577/hesoyam-for-claude-code.git
cd hesoyam-for-claude-code
./install.sh

Pick the profile that fits:

Or skip the wizard:

./install.sh --profile 1    # minimal
./install.sh --profile 2    # +memory
./install.sh --profile 3    # +knowledge
./install.sh --profile 4    # full

Obsidian not required. Profile 3 and 4 will ask if you use Obsidian. You can use a plain local markdown folder or skip vault setup entirely and just get the skills.

# Preview what will be installed (no changes made)
./install.sh --dry-run

# Or install pillar by pillar (legacy)
./install.sh --pillar orchestration
./install.sh --pillar config
./install.sh --pillar memory
./install.sh --pillar knowledge
./install.sh --pillar discovery

# Custom vault path
./install.sh --vault-path ~/my-obsidian-vault

See COMPATIBILITY.md for version requirements and token budget estimates.

Option 2: Plugin Marketplace (Individual Pillars)

# Inside Claude Code:

# Orchestration
/plugin marketplace add Yeachan-Heo/oh-my-claudecode
/plugin install oh-my-claudecode@omc

# Configuration & Security
/plugin marketplace add affaan-m/everything-claude-code
/plugin install everything-claude-code@everything-claude-code

# Memory
/plugin marketplace add thedotmack/claude-mem
/plugin install claude-mem

# Knowledge (Obsidian Skills)
/plugin marketplace add kepano/obsidian-skills
/plugin install obsidian@obsidian-skills

Option 3: Cherry-Pick What You Need

Read the pillar guides below. Take what serves you. Leave the rest. Come back when you’re ready.

---

💲 Pillar I — Orchestration (The $250K)

The Engine: oh-my-claudecode

“You don’t prompt anymore. You orchestrate.”

oh-my-claudecode transforms Claude Code from a single-agent terminal into a multi-agent execution engine with 32 specialized agents and 31+ skills. Workers spawn on demand and die when their task completes — zero idle resources.

Five Execution Modes:

Multi-Model Orchestration:

# Route architecture review to Codex, UI review to Gemini, Claude synthesizes
/ccg Review this PR — architecture (Codex) and UI components (Gemini)

HESOYAM Cheat Unlocked — Orchestration Playbooks:

hesoyam/
├── playbooks/
│   ├── sdlc-lifecycle.md          # ⭐ Full IT company SDLC (9 phases, end-to-end)
│   ├── greenfield-feature.md      # Interview → Plan → Swarm → Review
│   ├── legacy-migration.md        # Recon → Autopilot → Validate
│   ├── bug-hunt.md                # Deep Interview → Targeted Fix → TDD
│   ├── code-review-pipeline.md    # Parallel reviewers → Synthesis
│   └── sprint-execution.md        # Ultrapilot with daily checkpoints

Features People Actually Use Daily (see guides/daily-usage.md):

⚠️ Package naming gotcha: The repo and plugin are branded oh-my-claudecode, but the npm package is oh-my-claude-sisyphus. Upgrade via: npm i -g oh-my-claude-sisyphus@latest

---

🛡️ Pillar II — Configuration & Security (The Armor)

The Foundation: everything-claude-code

Battle-tested configs from an Anthropic hackathon winner, refined over 10+ months of daily production use.

ECC is the de facto standard for Claude Code configuration. 28 specialized agents. 125 skills. 60 commands. Works across Claude Code, Cursor, Codex, and OpenCode from a single repo.

What You Get:

• Agents — Planner, Architect, TDD Guide, Code Reviewer, Security Reviewer, Build Error Resolver, E2E Runner, Refactor Cleaner, Doc Updater, plus language-specific reviewers for TypeScript, Python, Go, Rust, Java, Kotlin, C++
• Security Hooks — Block --no-verify git flags, detect secrets in prompts (sk-, ghp_, AKIA), prevent agents from modifying linter configs instead of fixing code
• AgentShield — 1,282 tests and 102 static analysis rules. The --opus flag runs three Claude Opus agents in a red-team/blue-team/auditor pipeline
• Cross-Platform AGENTS.md — One config that deploys across Claude Code, Cursor, Codex, and OpenCode

Critical Token Optimization Rules (from the Longform Guide):

1. Don’t enable all MCPs at once. Your 200K context window can shrink to 70K with too many tools loaded. Keep under 10 MCPs enabled per project, under 80 tools active.
2. Convert MCPs to Skills where possible. Instead of the GitHub MCP eating context, create skills that use the gh CLI directly. Same functionality, freed context.
3. Disable auto-compact. Manually compact at logical intervals. Create a skill that suggests compaction based on defined criteria.
4. Use CLI flags to inject context dynamically. Instead of loading everything in CLAUDE.md every session, be surgical about what context loads when.

HESOYAM Cheat Unlocked — Stack-Specific Starter Configs:

hesoyam/
├── stacks/
│   ├── dotnet-clean-architecture/     # DDD, CQRS, EF Core, Result Pattern
│   ├── java-spring-boot/              # Spring Boot 3.3+, Hexagonal, JPA
│   ├── nextjs-fullstack/              # React, Tailwind, Prisma, tRPC
│   ├── python-django/                 # Django 5+, DRF, Celery
│   ├── python-fastapi/                # FastAPI, SQLAlchemy, Alembic
│   ├── ruby-rails/                    # Rails 7.2+, Hotwire, RSpec
│   ├── shopify-ecommerce/             # Liquid, Hydrogen, Shopify APIs
│   └── kubernetes-platform/           # Helm, ArgoCD, Crossplane

ECC v1.9.0 Features Most Guides Miss:

⚠️ Common gotcha: Do NOT add "hooks" to .claude-plugin/plugin.json. Claude Code v2.1+ auto-loads hooks. Declaring them causes “Duplicate hooks file” errors. See troubleshooting.

---

❤️ Pillar III — Memory & Persistence (The Health Bar)

“200K context window. Zero memory between sessions. You’re paying for a goldfish with a PhD.”

Primary: claude-mem

The most comprehensive memory system for Claude Code. Auto-captures everything Claude does during sessions, compresses it with AI, and injects relevant context back into future sessions.

• 5 Lifecycle Hooks: SessionStart → UserPromptSubmit → PostToolUse → Summary → SessionEnd
• SQLite + Full-Text Search — Everything stored locally at ~/.claude-mem/claude-mem.db
• Semantic Search — Natural language queries against your coding history
• Timeline Reports — Generate narrative journey reports from development history
• Multi-Mode Support — Code mode, law-study mode, email-investigation mode, and custom modes

# Install
/plugin marketplace add thedotmack/claude-mem
/plugin install claude-mem
# That's it. Context from previous sessions appears automatically.

Alternative: claude-brain (Single-File Memory)

For teams who want zero-dependency memory in a single portable .mv2 file:

/mind stats              # Memory statistics
/mind search "auth"      # Find past context
/mind ask "why JWT?"     # Ask your memory
/mind recent             # What happened lately

One file. Sub-millisecond search. Native Rust core. Grows ~1KB per memory. A year of use stays under 5MB.

Also Worth Exploring:

---

💾 Pillar IV — Knowledge & Second Brain (The Safe House)

The Vault: Obsidian + Claude Code

Claude Code’s context dies when the terminal closes. Your knowledge shouldn’t.

Obsidian Skills (by the CEO of Obsidian) — Drop-in agent skills that teach Claude Code to work natively with your Obsidian vault:

# Claude Code plugin install
/plugin marketplace add kepano/obsidian-skills
/plugin install obsidian@obsidian-skills

Claudesidian — A pre-configured Obsidian vault structure designed to work seamlessly with Claude Code:

claudesidian/
├── 00_Inbox/          # Temporary capture point
├── 01_Projects/       # Active initiatives
├── 02_Areas/          # Ongoing responsibilities
├── 03_Resources/      # Reference materials
├── 04_Archive/        # Completed work
├── 05_Attachments/    # Images, PDFs, files
└── 06_Metadata/       # Templates, config

Built-in commands: /thinking-partner, /daily-review, /research-assistant

Obsidian Claude Code MCP — Connect Claude Code to your vault via Model Context Protocol for real-time note access.

The Research Lab: NotebookLM Integration

notebooklm-py — Unofficial Python API for Google NotebookLM. Full programmatic access via Python, CLI, and AI agents:

pip install notebooklm-py

# Create research notebooks from Claude Code sessions
notebooklm create "Sprint 47 Research"
notebooklm source add "./architecture-decisions.md"
notebooklm source add "https://relevant-article.com"

# Generate deliverables
notebooklm generate audio "make it engaging" --wait
notebooklm generate slide-deck
notebooklm generate mind-map

# Ask questions across all sources
notebooklm ask "What are the key architectural tradeoffs?"

awesome-notebookLM-prompts (1.6K stars) — Field-tested slide prompts from researchers, founders, and designers. Turn brain dumps into presentation-ready decks.

awesome-notebooklm — Curated tools and examples: Podcastfy, Open NotebookLM, Zenmic, and more.

open-notebook — Self-hosted NotebookLM alternative. 16+ AI providers. Professional podcast generation. No vendor lock-in.

The Meeting Layer: Granola

Granola captures meeting context that feeds into your Obsidian vault and Claude Code sessions. The bridge between human conversations and agent workflows:

Meeting (Granola) → Structured Notes (Obsidian) → Context (Claude Code) → Action (Code)

HESOYAM Cheat Unlocked — Knowledge Pipeline:

hesoyam/
├── knowledge/
│   ├── obsidian-vault-template/       # PARA method + Claude Code ready
│   ├── meeting-to-code-pipeline.md    # Granola → Obsidian → Claude Code
│   ├── research-workflow.md           # NotebookLM + Claude Code research loop
│   ├── team-brain-architecture.md     # Shared knowledge across engineering team
│   └── daily-standup-automation.md    # Auto-generate standups from session history

🔗 Key Synergy: claude-mem + Obsidian (both optional, but together they’re magic)

See the full claude-mem + Obsidian Synergy Guide.

🔄 Git-Backed Obsidian — Version Control Your Brain

See the full Git-Backed Obsidian Workflow for:

• Auto-commit every 10 minutes via Obsidian Git plugin
• Claude Code SessionEnd hooks that commit vault changes
• Team vault structure with per-developer inboxes
• Conflict resolution strategies for shared vaults
• .gitignore template that protects secrets while sharing knowledge

Don’t use Obsidian? That’s fine. claude-mem alone gives you persistent memory. See Pick Your Path — Path A and Path B work without any external tools.

---

🗺️ Pillar V — Discovery & Ecosystem (The Map)

The Map: awesome-claude-code

The definitive catalog of the Claude Code ecosystem.

When you need to find a specific capability — a Kubernetes skill, a .NET reviewer agent, a Playwright E2E setup — this is where you look. Maintained by @hesreallyhim with editorial rigor (only Claude is allowed to submit PRs).

What it catalogs: Skills & slash commands, agent orchestrators, CLAUDE.md examples, CLI tools, output styles, tutorials, safety plugins, session restore tools, cost optimization guides, and domain-specific resources.

Also in the Discovery Layer:

Go deeper: Read the full Discovery & Ecosystem Guide for evaluation criteria, a monthly discovery workflow, and tool category breakdowns with token cost analysis.

---

🗺️ The Map: How the Cheats Stack

┌─────────────────────────────────────────────────────────────────────┐
│                         HESOYAM                                  │
│                                                                     │
│  ┌──────────────┐  ┌──────────────┐  ┌──────────────┐              │
│  │  Granola      │  │  Obsidian    │  │  NotebookLM  │   KNOWLEDGE │
│  │  (Meetings)   │─▶│  (Vault)     │◀─│  (Research)  │   LAYER     │
│  └──────┬───────┘  └──────┬───────┘  └──────────────┘              │
│         │                 │                                         │
│         ▼                 ▼                                         │
│  ┌──────────────────────────────────┐                               │
│  │         claude-mem / brain       │            MEMORY LAYER       │
│  │   (Persistent Session Memory)    │                               │
│  └──────────────┬───────────────────┘                               │
│                 │                                                    │
│                 ▼                                                    │
│  ┌──────────────────────────────────┐                               │
│  │    everything-claude-code        │            CONFIG LAYER       │
│  │  (Agents, Skills, Hooks, Rules)  │                               │
│  └──────────────┬───────────────────┘                               │
│                 │                                                    │
│                 ▼                                                    │
│  ┌──────────────────────────────────┐                               │
│  │     oh-my-claudecode             │            ORCHESTRATION      │
│  │  (Multi-Agent Execution Engine)  │            LAYER              │
│  └──────────────┬───────────────────┘                               │
│                 │                                                    │
│                 ▼                                                    │
│  ┌──────────────────────────────────┐                               │
│  │     awesome-claude-code          │            DISCOVERY          │
│  │  (Ecosystem Catalog & Updates)   │            LAYER              │
│  └──────────────────────────────────┘                               │
└─────────────────────────────────────────────────────────────────────┘

---

🧪 Recommended Loadouts

🟢 Solo Developer — “The Grove Street OG”

Obsidian Skills + claude-brain + ECC (selective install) + awesome-claude-code

Lightweight. One memory file. Strong guardrails. Discovery when needed.

🟡 Small Team (2-5 devs) — “The Heist Crew”

Obsidian + claude-mem + ECC (full install) + oh-my-claudecode (Swarm mode)

Shared memory. Parallel execution. Security hooks. Coordinated agents.

🔴 Engineering Org (5+ devs) — “The Vice City Empire”

Full HESOYAM Stack + Granola + NotebookLM + custom playbooks

Meeting capture → knowledge base → persistent memory → orchestrated execution → discovery of new capabilities. The full loop. You own the city.

🌟 Wanted Level — Setup Complexity

How deep do you want to go?

Pro tip: Start at 1 star. You can always type more cheat codes later.

---

🎯 Shipped Skills — The Cheat Sheet

HESOYAM ships with four ready-to-use skills (slash commands) in .claude/skills/:

Want to see all five pillars in action? Check the example CLAUDE.md for a complete real-world configuration.

---

🕹️ A Day in the Life (with HESOYAM)

flowchart LR
    subgraph MORNING["☀️ Morning"]
        A1["Open Claude Code"] --> A2["claude-mem auto-loads\nyesterday's context"]
        A2 --> A3["/daily-standup\ngenerates standup"]
    end

    subgraph WORK["💻 Working"]
        B1["Pick a ticket"] --> B2{Spec clear?}
        B2 -->|Yes| B3["Autopilot / Ralph Mode"]
        B2 -->|No| B4["Deep Interview first"]
        B4 --> B3
        B3 --> B5["ECC agents guard\nquality in real-time"]
        B5 --> B6["claude-mem captures\neverything silently"]
    end

    subgraph EOD["🌙 End of Day"]
        C1["PR reviewed by\nECC agents"] --> C2["Auto-commit +\npush to vault"]
        C2 --> C3["📱 Available on\nyour phone"]
    end

    subgraph FRIDAY["🍻 Friday"]
        D1["/promote-to-vault\ncurate the week"] --> D2["/ecosystem-check\nscan for new tools"]
    end

    MORNING --> WORK --> EOD --> FRIDAY

    style MORNING fill:#FFD700,color:#000
    style WORK fill:#FF6B35,color:#fff
    style EOD fill:#00D4AA,color:#fff
    style FRIDAY fill:#E040FB,color:#fff

---

📂 Repo Structure

hesoyam-for-claude-code/
├── README.md                          # You are here
├── CONTRIBUTING.md                    # How to contribute
├── COMPATIBILITY.md                   # Version matrix & token budgets
├── CODE_OF_CONDUCT.md                 # Community standards
├── LICENSE                            # MIT
├── install.sh                         # Unified installer (real automation)
│
├── .github/workflows/ci.yml          # Markdown lint + link check + structure validation
│
├── .claude/skills/                    # Shipped skills (slash commands)
│   ├── promote-to-vault/SKILL.md      # /promote-to-vault — curate mem → Obsidian
│   ├── daily-standup/SKILL.md         # /daily-standup — auto-generate standups
│   ├── research-sprint/SKILL.md       # /research-sprint — structured research
│   └── ecosystem-check/SKILL.md      # /ecosystem-check — monthly tool audit
│
├── playbooks/                         # Orchestration playbooks
│   ├── sdlc-lifecycle.md              # Complete IT company SDLC (9 phases)
│   ├── greenfield-feature.md
│   ├── legacy-migration.md
│   ├── bug-hunt.md
│   ├── code-review-pipeline.md
│   └── sprint-execution.md
│
├── stacks/                            # Stack-specific starter configs
│   ├── dotnet-clean-architecture/
│   ├── java-spring-boot/             # Spring Boot 3.3+, hexagonal architecture
│   ├── nextjs-fullstack/
│   ├── python-django/                # Django 5+, DRF, Celery
│   ├── python-fastapi/
│   ├── ruby-rails/                   # Rails 7.2+, Hotwire, RSpec
│   ├── shopify-ecommerce/
│   └── kubernetes-platform/
│
├── examples/                          # Real-world examples
│   └── CLAUDE.md.example              # Full HESOYAM config for a real project
│
├── knowledge/                         # Knowledge pipeline templates
│   ├── obsidian-vault-template/
│   ├── templates/                     # Research & decision templates
│   │   ├── research-brief.md
│   │   ├── technology-evaluation.md
│   │   ├── adr-template.md
│   │   └── spike-report.md
│   ├── meeting-to-code-pipeline.md
│   ├── research-workflow.md
│   ├── daily-standup-automation.md
│   └── team-brain-architecture.md
│
├── guides/                            # Deep-dive guides
│   ├── pick-your-path.md              # 🛤️ START HERE — choose your setup level
│   ├── daily-usage.md                 # What people actually use daily (80/20)
│   ├── discovery-ecosystem.md         # 🔭 Tool discovery & evaluation guide
│   ├── token-optimization.md
│   ├── memory-strategies.md
│   ├── multi-agent-patterns.md
│   ├── claude-mem-obsidian-synergy.md # 🔗 The killer combination
│   ├── obsidian-setup.md
│   ├── obsidian-git-workflow.md       # 🔄 Version control your brain
│   ├── notebooklm-workflows.md
│   └── security.md                    # 🔒 Protecting your workflow
│
└── community/                         # Community contributions
    ├── showcase/                       # Real-world setups
    │   ├── solopreneur-saas/          # Solo dev SaaS build (3 pillars)
    │   └── devops-duo/                # 2-person platform team (all 5 pillars)
    ├── recipes/                        # Specific workflow recipes
    └── troubleshooting/               # Common issues & fixes

---

🤝 Contributing

HESOYAM is a community cheat code. The more people contribute, the more powerful it gets.

We welcome contributions of all kinds:

What We Need Most

How to Contribute

1. Fork this repo
2. Create a branch (feature/my-awesome-playbook)
3. Add your contribution following the structure above
4. Open a PR with a clear description of what you’re adding and why
5. Get reviewed by the community

First-Time Contributors

Look for issues tagged good-first-issue. These are specifically chosen to be approachable:

• Add a new stack config for your favorite framework
• Write a recipe for a workflow you use daily
• Translate a guide into your language
• Fix a typo or improve clarity in existing docs

🌍 Translations Welcome

This repo aims to be accessible globally. We especially welcome translations into:

Hindi · Mandarin · Spanish · Japanese · Korean · Portuguese · German · French · Arabic · Russian

If you’re fluent in any of these (or any other language), please consider translating the core guides.

---

🙏 Credits & Acknowledgments

HESOYAM stands on the shoulders of giants. Every project referenced here represents hundreds (sometimes thousands) of hours of work by passionate developers. Please star the original repos.

---

⭐ Star History

If this repo helped you, smash that star button. It helps other developers find the cheat code.

Every star is a developer who stopped grinding and started winning.

---

📜 License

MIT — Use it, fork it, modify it, ship it. No wanted level. Just don’t forget to star the original repos.

---

Built by the Claude Code community “All you had to do was type the damn cheat code, CJ.”