{
  "markdown": "<div align=\"center\">\n\n![Hyperstack Banner](assets/banner.jpeg)\n\n**A disciplined engineering harness that forces AI agents to use ground-truth docs, precise designs, and programmatic verification.**\n\n<p>\n  <img src=\"https://img.shields.io/badge/license-MIT-blue?style=flat-square\" alt=\"MIT\" />\n  <img src=\"https://img.shields.io/badge/TypeScript-5-3178c6?style=flat-square&logo=typescript&logoColor=white\" alt=\"TypeScript\" />\n  <img src=\"https://img.shields.io/badge/Docker-ready-2496ED?style=flat-square&logo=docker&logoColor=white\" alt=\"Docker\" />\n  <img src=\"https://img.shields.io/badge/MCP-standard-6366f1?style=flat-square\" alt=\"MCP\" />\n</p>\n\n<p>\n  <img src=\"https://img.shields.io/badge/16_plugins-121_tools-6366f1?style=flat-square\" alt=\"Plugins\" />\n  <img src=\"https://img.shields.io/badge/29_skills-adversarial_gates-a855f7?style=flat-square\" alt=\"Skills\" />\n  <img src=\"https://img.shields.io/badge/SessionStart-hook_injected-f59e0b?style=flat-square\" alt=\"Hook\" />\n</p>\n\n</div>\n\n---\n\n## 🚀 What is Hyperstack?\n\n**Hyperstack is a disciplined engineering harness for AI coding agents.** \n\nIt provides the necessary **Ground Truth** (via 121 specialized MCP tools) and **Adversarial Enforcement** (via 29 rigorous skills) to transform a generic LLM into a high-precision Senior Engineer. \n\nUnlike standard \"polite\" instructions, Hyperstack uses **Iron Laws** and a **SessionStart hook** to force agents to check real documentation, follow precise design specs, and provide binary verification before shipping.\n\n## 🚀 Installation\n\n### 1. Recommended: Agentic (autopilot)\n\nThe fastest way to install Hyperstack is to let your AI agent do it for you. This works with **Cursor, Windsurf, Roo Code, Claude Code, or Gemini**. Simply paste this command:\n\n```text\nFetch and follow the instructions at https://raw.githubusercontent.com/orkait/hyperstack/main/install.md\n```\n\nThe autopilot will autonomously detect your environment, install the MCP server (Docker or Local), and **automatically link the Skills repository**. This is the only step required.\n\n---\n\n### 2. Manual Configuration (Advanced)\n\nIf you aren't using the Autopilot, follow the **Unified Bootstrap** to set up both the MCP server and the Skills:\n\n1.  **Clone & Initialize**:\n    ```bash\n    git clone https://github.com/orkait/hyperstack.git ~/.hyperstack\n    cd ~/.hyperstack && bun install\n    ```\n\n2.  **Run the Setup Script**:\n    ```bash\n    bun scripts/setup.ts\n    ```\n\n3.  **Follow the CLI**: The script auto-detects your IDE, generates the JSON patch, and outputs the symlink command for your skill directory.\n\n**Supported platforms** (verified April 2026, from official docs):\n\n| IDE / CLI | MCP Config Path | Skill Path | Schema |\n|---|---|---|---|\n| **Claude Code** | `~/.claude.json` | `~/.claude/skills/hyperstack` | JSON `mcpServers` |\n| **Gemini CLI** | `~/.gemini/settings.json` | - | JSON `mcpServers` |\n| **Qwen Code** | `~/.qwen/settings.json` | `~/.qwen/skills/hyperstack` | JSON `mcpServers` |\n| **Codex CLI** | `~/.codex/config.toml` | - | TOML `mcp_servers` |\n| **Cursor** | `~/.cursor/mcp.json` | `.cursor/rules/` (project) | JSON `mcpServers` |\n| **Windsurf** | `~/.codeium/windsurf/mcp_config.json` | - | JSON `mcpServers` |\n| **Kiro** | `~/.kiro/settings/mcp.json` | - | JSON `mcpServers` |\n| **Zed** | `~/.config/zed/settings.json` | - | JSON `context_servers` |\n| **VS Code + Copilot** | `~/.config/Code/User/mcp.json` | `.vscode/` (project) | JSON `mcpServers` |\n| **Roo Code** | `.roo/mcp.json` (project) | `.roo/rules/` (project) | JSON `mcpServers` |\n| **Cline** | `~/.config/Code/User/globalStorage/.../cline_mcp_settings.json` | - | JSON `mcpServers` |\n| **Continue.dev** | `.continue/mcpServers/mcp.json` (project) | - | JSON `mcpServers` |\n\n> [!TIP]\n> Run `bun scripts/setup.ts` and it will auto-detect your platform and output the exact patch - in the right format for your IDE. Codex users get TOML, everyone else gets JSON.\n\n## ⚡ The Manifesto\n\nIt is not just a library; it is a **disciplined harness** made of three tightly-coupled layers:\n\n1.  **The Harness**: Bootstraps the agent, routes internal specialist roles (`hyper`, `website-builder`), and enforces global development invariants.\n2.  **The MCP Ground Truth**: 16 TypeScript plugins (121 tools) that provide deterministic data. No hallucinated imports or invented component specs.\n3.  **The Adversarial Gates**: 29 skills with **\"Enforcement Teeth\"**. These aren't suggestions; they are Iron Laws supported by rationalization tables that counter every excuse an agent uses to skip quality gates.\n\n> [!IMPORTANT]\n> **Use Hyperstack if** you want to force your agent to check docs, write designs first, and prove its work.\n> **Skip Hyperstack if** you want frictionless autocomplete. We prioritize intentional friction that catches bugs before they ship.\n\n## 🏗️ Architecture\n\n```mermaid\ngraph TD\n    User([User Request]) --> Harness[Internal Harness]\n    Harness --> Gates{Adversarial Gates}\n\n    subgraph \"The Enforcement Layer\"\n        Gates -->|Iron Laws| Skills[29 Skills with Enforcement Teeth]\n        Skills -->|SessionStart| Injection[Auto-Context Injection]\n    end\n\n    subgraph \"The Knowledge Layer\"\n        Skills -->|Ground Truth| MCP[16 MCP Plugins / 121 Tools]\n        MCP -->|Designer| Design[DESIGN.md Pipeline]\n        MCP -->|Verified| Code[Programmatic Compliance Check]\n    end\n\n    Code -->|Verification Evidence| User\n\n    linkStyle 0 stroke:#9ca3af,stroke-width:2px\n    linkStyle 1 stroke:#9ca3af,stroke-width:2px\n    linkStyle 2 stroke:#9ca3af,stroke-width:2px\n    linkStyle 3 stroke:#9ca3af,stroke-width:2px\n    linkStyle 4 stroke:#9ca3af,stroke-width:2px\n    linkStyle 5 stroke:#9ca3af,stroke-width:2px\n    linkStyle 6 stroke:#9ca3af,stroke-width:2px\n    linkStyle 7 stroke:#9ca3af,stroke-width:2px\n```\n\n## 🤖 Agent Model\n\nIf you are an agent reading this, your mental model must be: **Process is Priority.** You are operating within a system that requires deterministic grounding and formal design before implementation.\n\n- **Conductance**: `hyper` owners classification and gate verification.\n- **Specialization**: `website-builder` handles visual and frontend implementation.\n- **Persistence**: You run in a stable Docker container with a persistent lifecycle.\n\n---\n\n## 🚀 Quickstart\n\n### 🐳 Docker (Default)\n\nHyperstack uses a persistent container plus `docker exec`. This keeps startup cheap across sessions and ensures 100% environment stability.\n\n1. Pull the image:\n\n```bash\ndocker pull ghcr.io/orkait/hyperstack:main\n```\n\n2. Start the persistent container:\n\n```bash\ndocker rm -f hyperstack-mcp 2>/dev/null\ndocker run -d --name hyperstack-mcp --restart unless-stopped \\\n  --memory=512m --cpus=1 \\\n  --entrypoint sleep \\\n  ghcr.io/orkait/hyperstack:main infinity\n```\n\n3. Add this to your MCP settings (`~/.claude.json`, Cursor, Windsurf, etc.):\n\n```json\n{\n  \"mcpServers\": {\n    \"hyperstack\": {\n      \"command\": \"docker\",\n      \"args\": [\"exec\", \"-i\", \"hyperstack-mcp\", \"bun\", \"/app/src/index.ts\"]\n    }\n  }\n}\n```\n\n### 🤖 Agent Autopilot\n\nIf you are using Claude Code, Cursor, Windsurf, Roo Code, or Gemini, you can use the autopilot to self-configure. Paste this at your agent:\n\n```text\nFetch and follow the instructions at https://raw.githubusercontent.com/orkait/hyperstack/main/install.md\n```\nThe autopilot will detect your environment and propose the correct Docker-based configuration block.\n\n\n\n### 🔧 Install the skills\n\nThe MCP server gives you tools. The skills give you discipline. Install both:\n\n```bash\ngit clone https://github.com/orkait/hyperstack.git ~/.claude/skills/hyperstack\n```\n\nAfter installing, the SessionStart hook (at `hooks/session-start.mjs`) will auto-inject the `hyperstack` skill into every session. No manual activation needed.\n\n### 💻 From source\n\n```bash\ngit clone https://github.com/orkait/hyperstack.git\ncd hyperstack\nbun install\nbun bin/hyperstack.mjs   # same entrypoint the published bin uses\nbun start                # no build step\nbun dev                  # watch mode\nbun run build            # tsc --noEmit (type-check only, no dist output)\n```\n\nNode 18+ required.\n\n---\n\n## 🧠 The Four-Layer Framework\n\nOne framework, four layers: **Ground Truth** (MCP), **Enforcement** (Skills), **Orchestration** (Roles), and **Personas** (domain lenses, auto-engaged by trigger). Each persona binds its plugin, skills, and routing through one manifest.\n\n### Layer 1: MCP Plugins (Ground Truth)\n\nYour AI calls these for deterministic data. Memory is not acceptable. Every plugin serves curated TypeScript data and architectural patterns.\n\n| Category | Plugins | Domain Coverage |\n|---|---|---|\n| 🛠️ **System** | `hyperstack` | Autonomous Environment Detection, MCP Configuration Patching, Lifecycle |\n| 🎨 **UI Engine** | `designer`, `design-tokens`, `ui-ux`, `shadcn` | Design Systems, OKLCH, Typography, Accessibility, Component Specs |\n| ⚛️ **Frontend** | `react`, `reactflow`, `motion`, `lenis` | Next.js 15, RSC, Animation Curves, Smooth Scroll, DAG Layouts |\n| 🐹 **Backend** | `echo`, `golang`, `rust` | Professional Go Recipes, Rust Borrow Checker patterns, Clean Architecture |\n| 📈 **Product & Growth** | `optimizer`, `product-manager`, `marketing`, `reflect` | DSA technique matching, Product risk gates (JTBD, RICE), Positioning & GTM, Target-customer screen/feature review |\n\n> [!TIP]\n> **121 tools across 16 plugins**. Every tool is designed to provide the \"Senior Engineer\" answer, bypassing the \"AI Slop\" default.\n\n### Layer 2: Skills (Enforcement Teeth)\n\nMarkdown with adversarial enforcement. Each skill contains an **Iron Law** that the agent is bound to follow.\n\n> [!CAUTION]\n> ### ⚖️ The Iron Laws of Hyperstack\n> - **NO CODE** without MCP grounding.\n> - **NO VISUAL CODE** without an approved `DESIGN.md`.\n> - **NO COMPLETION CLAIMS** without programmatic verification evidence.\n> - **NO REFACTOR** without a failing test first.\n> - **NO PATTERN** without a named Force.\n\nThese laws are backed by **Rationalization Tables**-pre-written counters to every excuse an AI agent uses to skip quality gates.\n\n### Layer 3: Agents (Orchestration & Routing)\n\nThe internal harness is what ties the public layers together by managing process and domains:\n\n- bootstrap is injected at session start from generated runtime context\n- `hyper` owns classification, routing, gates, and verification\n- `website-builder` specializes in website-facing design and implementation work\n- roles are internal and auto-called, not user-invoked commands\n\n### Layer 4: Personas (Domain Lenses)\n\nDomain-expert lenses `hyper` auto-engages when their triggers match the request - the user never has to name them. Each binds an MCP plugin + skills + routing via `personas/<id>/persona.json`, and the trigger table is compiled into the session bootstrap from those manifests (single source of truth).\n\n| Persona | Mode | Lens |\n|---|---|---|\n| `product-manager` | gate | blocks net-new builds until value + viability risks carry evidence |\n| `bro` | capability | generalist execution: decomposition, evidence discipline, code quality |\n| `marketing` | capability | positioning, messaging, copy, brand voice, GTM |\n| `reflect` | capability | reviews screens and features AS the real target customer (8 archetypes + panel mode) |\n\n<details>\n<summary><strong>🧱 Core (13)</strong> - workflow, discipline, gates used on every task</summary>\n\n| Skill | Role |\n|---|---|\n| `blueprint` | Hard gate: no code without an approved design |\n| `forge-plan` | MCP-verified task-by-task implementation plan |\n| `run-plan` | Execute an existing plan |\n| `best-practices` | 8-step Senior SDE framework, 5 Iron Laws, patterns, coupling, review and ops rules |\n| `ship-gate` | No completion claims without fresh verification evidence |\n| `deliver` | Final verification and delivery |\n| `test-first` | No production code without a failing test first |\n| `debug-discipline` | Root cause first, 3-strike escalation |\n| `code-review` | Dispatch reviewer subagent, handle feedback technically |\n| `autonomous-mode` | Full end-to-end execution, only stops on failure |\n| `subagent-ops` | Fresh agent per task, two-stage review |\n| `parallel-dispatch` | Concurrent agent dispatch for independent tasks |\n| `worktree-isolation` | Clean workspace isolation before feature work |\n\n</details>\n\n<details>\n<summary><strong>🎯 Domain (7)</strong> - specialized skills for specific contexts</summary>\n\n| Skill | Role |\n|---|---|\n| `designer` | Intention gate - produces DESIGN.md contract before any visual code |\n| `shadcn-expert` | shadcn/ui Base UI architect - ONLY when user picks shadcn in designer Q11b |\n| `react-pro-coder` | 56 React/Next.js rules with stable IDs - rendering, state placement, audits |\n| `behaviour-analysis` | UI/UX state audits, Nielsen heuristics, interaction matrices |\n| `security-review` | OWASP audits, vulnerability checklists |\n| `readme-writer` | Evidence-based README generation (this skill) |\n| `python-pro-coder` | 60 FastAPI/Pydantic v2/SQLAlchemy 2.0 rules - layering, schemas, async, dead deps |\n\n</details>\n\n<details>\n<summary><strong>🔭 Meta (2)</strong> - skills about skills</summary>\n\n| Skill | Role |\n|---|---|\n| `hyperstack` | Force-injected at session start via hook - the enforcement payload |\n| `testing-skills` | RED-GREEN-REFACTOR pressure testing for skills using subagents |\n\n</details>\n\nFull index at `skills/INDEX.md`. Regenerate with `bun scripts/generate-skills-index.ts` after adding or editing any skill.\n\n---\n\n## 🔒 Adversarial Enforcement\n\nOrdinary skill markdown is a polite suggestion. Polite suggestion fails when an AI model is under pressure to \"be helpful fast.\" Hyperstack skills are written adversarially:\n\n- **1% Rule**: If there is even a 1% chance a skill applies, the agent **must** invoke it.\n- **Rationalization Tables**: We have already written down every excuse your AI will use to skip a gate, with a firm technical counter for each.\n- **Loophole Closure**: The \"Spirit of the Law\" is explicitly defined as the \"Letter of the Law\" to prevent shortcut-hunting.\n\n\n---\n\n## 🎨 The designer agent \n\nWhen you say, **“build me a SaaS dashboard”**:\n\n1. **SessionStart** already puts in `hyperstack`, so AI know system is there.\n2. **Blueprint skill** sees visual job and sends it to `hyperstack:designer`.\n3. **Designer skill** runs `designer_resolve_intent(product)` to guess industry, personality, style, density, and mode.\n4. Designer asks **3 questions** in base mode, or **12 questions** in advanced mode.\n5. Like **Q11b** will ask what component library to use: shadcn, raw Tailwind, MUI, Mantine, Chakra, Ant Design, or custom.\n6. Designer makes a **DESIGN.md** contract with 10 parts: theme, colors, type, spacing, components, motion, elevation, do/don’ts, responsive rules, and anti-patterns.\n7. User approves the **DESIGN.md**.\n8. **Forge-plan** reads it and makes one task for each section. If user picked shadcn, it calls `shadcn_get_component`. If not, it builds from the DESIGN.md spec.\n9. Build tasks run with MCP tools as ground truth.\n10. **designer_verify_implementation** checks build against **DESIGN.md**.\n11. **Ship-gate** blocks final completion unless build passes the **DESIGN.md** rules.\n\nAI cannot jump ahead. Every step has hard gate. Excuses already blocked by rationalization tables.\n\n\n---\n\n## 🛠️ Available Tools\n\n### 📦 Supported Versions\n\nOpinionated stack support for the latest stable stable releases. This prevents \"AI Slop\" by enforcing modern patterns (e.g., React 19 Actions, Tailwind v4 tokens).\n\n| Technology | Supported Version | Role |\n| :--- | :--- | :--- |\n| **React** | 19.x | Core Library |\n| **Next.js** | 15.x | Application Framework |\n| **Tailwind CSS** | v4.x | Design Tokens & Constraints |\n| **Motion** | 12.x (fka Framer Motion) | Orchestrated Animations |\n| **React Flow** | 12.x | Node-based Systems |\n| **Lenis** | 1.1.x+ | Smooth Scroll Engine |\n| **Zustand** | 5.x | State Management |\n| **shadcn/ui** | Base UI Edition | Component Primitive Patterns |\n| **Echo (Go)** | v4.x | Backend Framework |\n| **Go** | 1.22+ | Backend Language |\n| **Rust** | 1.77+ | Systems Language |\n| **Bun** | 1.1.x+ | Runtime Environment |\n\n\n<details>\n<summary><strong>🎨 Designer</strong> - <code>designer_*</code> (19 tools)</summary>\n\n- `designer_resolve_intent` - Auto-detect industry, personality, style from product description\n- `designer_list_personalities` + `designer_get_personality` - 6 personality clusters from 58 real company design systems\n- `designer_list_presets` + `designer_get_preset` - 9 code-ready token presets (Linear, Stripe, Vercel, Apple, Carbon, shadcn, Notion, Supabase, Figma)\n- `designer_get_industry_rules` - 15 industry profiles with must-have/never-use constraints\n- `designer_get_cognitive_law` - 11 laws (Fitts, Hick, Miller, Gestalt, Von Restorff, Serial Position, F-Pattern, Z-Pattern, Jakob, Doherty, Peak-End)\n- `designer_get_page_template` - 13 page types with section anatomy\n- `designer_get_composition_rules` - Visual hierarchy, CRAP, whitespace, fold, reading patterns\n- `designer_get_interaction_pattern` - Form design, navigation, empty states, micro-interactions\n- `designer_get_ux_writing` - Button labels, error messages, confirmation dialogs\n- `designer_get_landing_pattern` - Hero, social proof, pricing, CTA optimization\n- `designer_get_design_system` - Specific values from 7 premium systems\n- `designer_get_font_pairing` - 21 curated pairings with Google Fonts imports\n- `designer_get_anti_patterns` - 50+ anti-patterns (the AI slop fingerprint) filterable by category/industry\n- `designer_search` - Cross-domain search\n- `designer_generate_design_brief` - Assemble structured brief\n- `designer_generate_implementation_plan` - Parse DESIGN.md into executable task list with MCP calls per section\n- `designer_verify_implementation` - Programmatic compliance check against DESIGN.md\n</details>\n\n<details>\n<summary><strong>🧩 shadcn/ui</strong> - <code>shadcn_*</code> (5 tools, optional)</summary>\n\nOnly invoked when the user explicitly chose shadcn in designer Q11b.\n\n- `shadcn_get_rules` - Architectural rules and mandatory checklist (call first)\n- `shadcn_list_components` - Curated component catalog\n- `shadcn_get_component` - Full spec: primitive, data-slots, variants, sizes\n- `shadcn_get_snippet` - Canonical usage example\n- `shadcn_get_composition` - Which components compose for a page type (bridge from designer page templates)\n</details>\n\n<details>\n<summary><strong>✨ Design Tokens</strong> - <code>design_tokens_*</code> (7 tools)</summary>\n\n- `design_tokens_list_categories` - 10 token categories (colors, spacing, grid, radius, shadows, motion, z-index, opacity, component sizing, typography)\n- `design_tokens_get_category` - CSS, rules, gotchas per category\n- `design_tokens_get_color_ramp` - OKLCH values + semantic roles\n- `design_tokens_get_procedure` - 8 step-by-step build procedures\n- `design_tokens_get_gotchas` - Aggregate implementation mistakes\n- `design_tokens_generate` - Complete Tailwind v4 token file from a palette\n- `design_tokens_search` - Cross-category search\n</details>\n\n<details>\n<summary><strong>💅 UI/UX Principles</strong> - <code>ui_ux_*</code> (6 tools)</summary>\n\n- `ui_ux_list_principles` - Browse by domain (typography, color, accessibility, responsive, motion)\n- `ui_ux_get_principle` - Rule, detail, CSS examples, anti-patterns\n- `ui_ux_get_component_pattern` - Button, card, badge, form specs\n- `ui_ux_get_checklist` - Pre-ship checklist per domain\n- `ui_ux_get_gotchas` - Common UI mistakes and fixes\n- `ui_ux_search` - Cross-domain search\n</details>\n\n<details>\n<summary><strong>⚛️ React Flow</strong> - <code>reactflow_*</code> (9 tools)</summary>\n\n- `reactflow_list_apis` - Browse 56 APIs by kind\n- `reactflow_get_api` - Full reference: props, usage, tips\n- `reactflow_search_docs` - Full-text search\n- `reactflow_get_examples` - Curated code examples by category\n- `reactflow_get_pattern` - Enterprise patterns (zustand-store, drag-and-drop, SSR)\n- `reactflow_get_template` - Production-ready starters\n- `reactflow_get_migration_guide` - v11 to v12 breaking changes\n- `reactflow_generate_flow` - Generate a flow from prose\n</details>\n\n<details>\n<summary><strong>🎬 Motion for React</strong> - <code>motion_*</code> (7 tools)</summary>\n\n- `motion_list_apis` - Browse 33 APIs\n- `motion_get_api` - Full reference with props and usage\n- `motion_search_docs` - Full-text search\n- `motion_get_examples` - Animation examples by category (gestures, scroll, layout)\n- `motion_get_transitions` - Transition reference for tween, spring, inertia\n- `motion_generate_animation` - Generate animation snippet from description\n- `motion_cheatsheet` - Quick reference\n</details>\n\n<details>\n<summary><strong>🌊 Lenis</strong> - <code>lenis_*</code> (6 tools)</summary>\n\n- `lenis_list_apis` - Options, methods, events\n- `lenis_get_api` - Full reference with snippets\n- `lenis_get_pattern` - Next.js, GSAP, Framer Motion integrations\n- `lenis_generate_setup` - Complete Lenis setup\n- `lenis_cheatsheet` - Required CSS and pitfalls\n- `lenis_search_docs` - Full-text search\n</details>\n\n<details>\n<summary><strong>⚛️ React + Next.js</strong> - <code>react_*</code> (4 tools)</summary>\n\n- `react_list_patterns` - All React/Next.js patterns\n- `react_get_pattern` - Full implementation with anti-patterns\n- `react_get_constraints` - Hard rules (e.g., no `useEffect` for fetching)\n- `react_search_docs` - Search patterns and rules\n</details>\n\n<details>\n<summary><strong>🐹 Echo (Go)</strong> - <code>echo_*</code> (6 tools)</summary>\n\n- `echo_list_recipes` - Browse 19 recipes\n- `echo_get_recipe` - Full recipe (jwt-auth, websocket, sse)\n- `echo_list_middleware` + `echo_get_middleware` - 13 middleware components\n- `echo_decision_matrix` - Echo vs standard library\n- `echo_search_docs` - Full-text search\n</details>\n\n<details>\n<summary><strong>🐹 Golang</strong> - <code>golang_*</code> (6 tools)</summary>\n\n- `golang_list_practices` - Browse 18 best practices\n- `golang_get_practice` - Rule, reasoning, good/bad examples\n- `golang_list_patterns` + `golang_get_pattern` - 10 Go-idiomatic design patterns\n- `golang_get_antipatterns` - Common mistakes and fixes\n- `golang_search_docs` - Search practices and patterns\n</details>\n\n<details>\n<summary><strong>🦀 Rust</strong> - <code>rust_*</code> (4 tools)</summary>\n\n- `rust_list_practices` + `rust_get_practice` - 18 best practices\n- `rust_cheatsheet` - Ownership rules, pointer types, performance\n- `rust_search_docs` - Search all practices\n</details>\n\n---\n\n## 🤝 Contributing\n\nWe welcome contributions that follow the **Disciplined Engineering** standard.\n\n1.  **Plugins**: Must follow the `index.ts` + `data.ts` + `tools/` + `snippets/` pattern.\n2.  **Skills**: Must include `category` frontmatter and adhere to the Adversarial Enforcement style.\n3.  **Verification**: All PRs must pass the full `npm run build` (Type-check) and CI suite.\n\n```bash\n# Regenerate the skills index after editing\nbun scripts/generate-skills-index.ts\n```\n\n## 📄 License\n\nMIT © [Orkait](https://github.com/orkait) | Adversarial philosophy inspired by [Jesse Vincent's Superpowers](https://github.com/obra/superpowers).\n\n---\n\n## 🙏 Acknowledgements\n\nThe enforcement philosophy behind Hyperstack's gate skills - Iron Laws, 1% Rule, rationalization tables - was adopted from [obra/superpowers](https://github.com/obra/superpowers) (MIT © Jesse Vincent). We agreed with how it frames AI compliance: adversarially, not politely. See [CREDITS.md](./CREDITS.md).\n",
  "bytes": 23458,
  "sha": "90ff1223237f38c9ac25c18babb9f9dfaeeb2910cc7e6dd0ba845c797365af25",
  "repo_slug": "orkait/hyperstack",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_orkait_hyperstack_892fa3e7/readme"
}