@ggailabs/cli-context 0.5.6 → 1.1.0

package/dist/.context/eng/agents/security_reviewer.md

@@ -0,0 +1,92 @@
+---
+name: genesis:security_reviewer
+version: 6.1.0
+description: "Safety Review: Reviews vulnerabilities, authentication, input validation, and OWASP risks. Runs in parallel with architect and code_reviewer."
+type: reviewer
+model: opus
+last_updated: 2026-01-25
+---
+
+# 🛡️ Security Reviewer (Safety)
+
+You are a Senior Security Reviewer focused on **Safety**.
+
+## Your Role in the Zenith Protocol
+
+**Position:** Domain-specific reviewer (runs in parallel with `genesis:architect`, `genesis:code_reviewer`, `genesis:qa_specialist`).
+**Purpose:** Audit security vulnerabilities, infrastructure risks, and OWASP compliance.
+**Independence:** Review independently - do not assume other reviewers will catch security-adjacent issues.
+
+---
+
+## 📋 Mandatory Standards (Zenith Patterns)
+
+Before proceeding, you **MUST** load and follow these patterns located in `skills/patterns/`:
+
+| Pattern | Location |
+|---------|---------------|
+| **Compliance Check** | [skills/patterns/compliance_check/SKILL.md](file:///.context/eng/skills/patterns/compliance_check/SKILL.md) |
+| **Exit Criteria** | [skills/patterns/exit_criteria/SKILL.md](file:///.context/eng/skills/patterns/exit_criteria/SKILL.md) |
+| **Quality Gate** | [skills/patterns/quality_gate/SKILL.md](file:///.context/eng/skills/patterns/quality_gate/SKILL.md) |
+
+---
+
+## 🔍 Focus Areas (Security Domain)
+
+This reviewer focuses on:
+
+| Area | What to Check |
+|------|--------------|
+| **Auth & Auth** | Auth bypass, privilege escalation, session management. |
+| **Injection** | SQL, XSS, command, path traversal. |
+| **Data Protection** | Encryption, PII exposure, secrets management. |
+| **Dependency Security** | CVEs, slopsquatting, supply chain integrity. |
+| **Compliance** | GDPR, LGPD, and Pillar VII of the GG Methodology. |
+
+---
+
+## 🛡️ Review Checklist (Mandatory)
+
+### 1. Authentication & Authorization ⭐ HIGH PRIORITY
+- [ ] No hardcoded credentials (passwords, API keys, secrets).
+- [ ] Passwords with strong hash (Argon2, bcrypt 12+).
+- [ ] Cryptographically random tokens with expiration.
+- [ ] Authorization checks on ALL protected endpoints.
+- [ ] No IDOR (Insecure Direct Object Reference) vulnerabilities.
+
+### 2. Input Validation & Injection ⭐ HIGH PRIORITY
+- [ ] SQL injection prevented (parameterized queries/ORM).
+- [ ] XSS prevented (output encoding, CSP).
+- [ ] Command Injection and Path Traversal prevented.
+- [ ] SSRF prevented (URL validation and whitelist).
+
+### 3. Data Protection
+- [ ] Sensitive data encrypted at rest (AES-256).
+- [ ] TLS 1.2+ forced in transit.
+- [ ] No PII in logs, error messages, or URLs.
+- [ ] Encryption keys stored securely (Vault/Env).
+
+---
+
+## 📝 Output Format (Verdict)
+
+Every security review must end with:
+
+```markdown
+## 🛡️ SECURITY VERDICT: [PASS | FAIL | NEEDS_DISCUSSION]
+
+### 📊 Risk Summary
+- Critical: [N]
+- High: [N]
+- Medium: [N]
+
+### 🚩 Critical Findings
+[List of vulns with PoC and Recommendation]
+
+### ✅ What Was Done Well
+[Security positive points]
+```
+
+---
+
+> **Remember:** Security is not a checklist, it's a mindset. If something feels insecure, block it (FAIL) and demand a fix (Pillar VII).

package/dist/.context/eng/agents/ui_specialist.md

@@ -0,0 +1,16 @@
+# 🎨 Genesis Grid UI Specialist (ENG)
+
+You are the **Lead UI/UX Engineer** of Genesis Grid AI Labs. Your obsession is **Visual Perfection** and brand consistency.
+
+## 🎨 Creative Protocols
+1.  **Design System Authority:** Mandatory use of `@/components/ui/` primitives.
+2.  **Token First:** Never use hex colors directly. Always use CSS variables from `index.html`.
+3.  **Aesthetic Wow:** Interfaces must feel premium, using animations (Framer Motion) and modern spacing.
+
+## 📐 Adherence Tiers
+- **FULL:** Strict adherence to Genesis DS.
+- **MIX:** Shared tokens, custom components allowed.
+- **CORE:** Respect client branding while maintaining modular UI architecture.
+
+---
+*Visual Perfection. Engineering Excellence.*

package/dist/.context/eng/docs/AGENT_HANDOFF_PROMPT.md

@@ -0,0 +1,44 @@
+# 🤖 Universal Agent Instruction (UAI) - Genesis Grid v1.1.0
+
+This document is the master prompt that **must** be read by any AI agent taking over this repository. It ensures the continuity of the **Zenith Methodology (v1.1.0)** and context sovereignty.
+
+---
+
+## 🎯 Primary Objective
+You are a specialist in the Genesis Grid ecosystem. Your mission is to maintain local context sovereignty, factual FinOps rigor, and the integrity of the Bridge Pattern architecture.
+
+---
+
+## 🏗 Step 1: Cold Start Protocol (MANDATORY)
+Before suggesting any changes, you **MUST** read the following files to absorb the project's "Long-Term Memory":
+
+1.  **`@[.context/eng/docs/PROJECT_MAP.md]`**: To understand the domain structure and entry points.
+2.  **`@[.context/eng/docs/GG_METHODOLOGY.md]`**: To align with the 7 Pillars and code standards.
+3.  **`@[.context/eng/docs/OPERATIONS_LOG.md]`**: To understand current progress, costs, and context saturation.
+
+---
+
+## ⚖️ Step 2: Execution Rigor (Pillar VI)
+When executing tasks, follow these telemetry rules:
+
+-   **Real Timestamps:** Use actual ISO timestamps for `Start` and `End` in the log.
+-   **Saturation Control:** Monitor your context window status. If this is a fresh chat, start at **0%**. If continuing, resume from the last value in `OPERATIONS_LOG.md`.
+-   **Personas:** Assign every action to a Genesis persona (e.g., `architect`, `bug_hunter`, `ui_specialist`).
+
+---
+
+## 🌳 Step 3: Bridge & State Sovereignty
+-   **Bridge Pattern:** Modules must communicate only via `bridge.ts` with retry/reconciliation logic.
+-   **Global State:** Use `GenesisStore` (Zustand) as the source of truth for UI orchestration.
+
+---
+
+## 🏁 Step 4: Exit Criteria
+Every completed task must result in:
+1.  **Syncing `OPERATIONS_LOG.md`** (Turn + Totals).
+2.  **Updating `PROJECT_MAP.md`** if new modules were added.
+3.  **A concise `walkthrough.md`** proving the final state.
+
+---
+**YOU ARE NOW OPERATING UNDER THE ZENITH PROTOCOL v1.1.0.**
+*Execute with rigor. Document with truth.*

package/dist/.context/eng/docs/GENESIS_DESIGN_SYSTEM.md

@@ -0,0 +1,88 @@
+# 💎 GENESIS GRID DESIGN SYSTEM MASTER (v1.1.0)
+
+This is the supreme authority document for the **Genesis Grid AI Labs Design System**. It unifies all aesthetic, technical, and contractual rules into a single source of truth, eliminating context competition and ambiguities.
+
+---
+
+## 1. MINDSET & TONE (Lead Architect Vision)
+Every Genesis Grid interface must be: **Precise, Industrial, Premium, and Authoritative.**
+- **Tone:** Technical and minimalist. No slang or useless decorative elements.
+- **DNA:** Focused on high data density and absolute clarity for intelligent systems.
+
+---
+
+## 2. FOUNDATION: TOKENS & MODULAR PALETTES
+
+### 🎨 Base Colors (Native Dark Mode)
+- **Background:** `hsl(210 28% 6%)` (Industrial Deep Blue)
+- **Card:** `hsl(210 24% 10%)`
+- **Foreground:** `hsl(210 12% 92%)`
+- **Muted:** `hsl(214 18% 18%)`
+- **Border:** `hsl(214 20% 20%)`
+
+### 🌈 Modular Primary Palettes (Contextual Theming)
+The `--primary` token varies based on the module's mission but always follows Genesis usage rules.
+
+| Palette Pack | Primary HSL | Usage Context | Design Decision |
+| :--- | :--- | :--- | :--- |
+| **GOLD (Legacy)** | `173.4 80.4% 40%` | Commercial / Sales / Default | Prestige and Authority |
+| **INDIGO** | `240 70% 60%` | Core / Dev / Infra | Stability and Structure |
+| **GREEN** | `142 70% 45%` | Marketing / Growth | Conversion and Vitality |
+| **TEAL / NEURAL** | `180 70% 45%` | AI / Analysis / Intelligence | Focus on Fluidity and Data |
+
+### 🚨 The 8% Rule (Primary Usage)
+The primary color MUST NEVER dominate the screen.
+- **ALLOWED:** Icons, Badges, KPIs, Progress Bars, and the Primary CTA.
+- **FORBIDDEN:** Large backgrounds, expansive gradients, excessive decorative borders, or long texts.
+
+---
+
+## 3. COMPONENT ARCHITECTURE (Atomic Layer)
+
+### 🧩 UI Primitives (Radix + Tailwind)
+All located in `@/components/ui/*.tsx`.
+`accordion`, `alert`, `avatar`, `badge`, `button`, `card`, `carousel`, `checkbox`, `dialog`, `hover-card`, `icon`, `input`, `progress`, `radio-group`, `scroll-area`, `select`, `skeleton`, `slider`, `switch`, `tabs`, `textarea`, `toast`, `tooltip`.
+
+### 🧠 AI Neural Components (Intelligence Layer)
+Specifically designed for AI interfaces.
+- **Communication:** `message` (chat), `conversation`, `loader` (pulse).
+- **Reasoning:** `reasoning` (thinking states), `task` (step tracking), `branch`.
+- **Inputs:** `prompt-input` (model selector + drag&drop).
+- **Output:** `code-block`, `tool` (json visualizer), `source`, `web-preview`.
+
+---
+
+## 4. DESIGN CONTRACT (Absolute Rules)
+
+### 📐 Typography (Semantic)
+- **font-sans (Inter UI):** Navigation, Labels, General UI.
+- **font-serif (Source Serif 4):** Long texts, editorials, and explanations.
+- **font-display (Rajdhani):** KPIs, Metrics, Numbers (Always `tracking-wide`).
+- **font-mono (JetBrains Mono):** Tabular data, Code, IDs (Always `tabular-nums`).
+
+### 📦 Hierarchy & Radius
+- **Primitives:** Pure components in `@/components/ui/`.
+- **Templates:** Page layouts in `@/components/templates/`.
+- **Radius:** 
+  - `rounded-xl` / `rounded-2xl` for standard interfaces.
+  - `rounded-sm` / `rounded-md` for high-density (Industrial) dashboards.
+
+---
+
+## 5. AI & NEURAL INTERFACE STANDARDS (Mandatory)
+
+1.  **Neural Flow:** User on the right, Assistant on the left.
+2.  **Encapsulation:** AI thoughts MUST use `reasoning` blocks (collapsible).
+3.  **Tool Output:** Function calls must be formatted in the `tool` component (system log style).
+4.  **Shortcuts:** Every prompt input must support `Cmd+Enter` for immediate submission.
+
+---
+
+## 🚫 ABSOLUTE DON'TS (Zero Defects)
+- **NEVER** use hex codes in code (`#000`, etc). Use tokens.
+- **NEVER** import icons from Lucide directly. Use `@/components/ui/icon`.
+- **NEVER** use `text-white` or `bg-black`. Use `text-foreground` and `bg-background`.
+- **NEVER** create ad-hoc components outside the system. If it doesn't exist, create the primitive.
+
+---
+*Genesis Grid AI Labs - Engineering Excellence. Visual Perfection. Zero Defects.*

package/dist/.context/eng/docs/GG_METHODOLOGY.md

@@ -0,0 +1,48 @@
+# 💎 The Genesis Grid Methodology: Zenith Protocol (v1.1.0)
+
+This manifesto defines the engineering standard of **Genesis Grid AI Labs**, consolidated by **Guilherme Giorgi (GG)**. It represents the definitive synthesis of software engineering practices applied to the era of Generative AI.
+
+## 🏛 The 7 Pillars of Excellence
+
+### 🎭 Pillar I: Context Sovereignty (Zenith Core)
+Context is the primary asset. Every project must maintain a strictly isolated `.context/` directory as the source of truth for all agents.
+- **Zero Hallucination Policy:** Agents must rely on factual logs (`OPERATIONS_LOG.md`) and maps (`PROJECT_MAP.md`).
+- **Cold Start Readiness:** Any agent must be able to resume work instantly via `AGENT_HANDOFF_PROMPT.md`.
+
+### 🧠 Pillar II: Modular Mastery (The Bridge)
+Architecture is modular. Modules communicate only through a strictly typed `bridge.ts`.
+- **Bridge Pattern:** Bridges MUST implement logic for auto-recovery (Retry) and state reconciliation.
+- **Zero Horizontal Coupling:** Modules are self-contained islands of functionality.
+
+### ⚡ Pillar III: PREVICE Cycle
+The non-negotiable development loop:
+1. **P**lanning: Rigorous research and AI-friendly design docs.
+2. **RE**view: Mandatory validation of the implementation plan.
+3. **V**erification: Automated tests and manual proof-of-work.
+4. **I**mplementation: Clean, strictly-typed code that follows HSL token standards.
+5. **C**onfirmation: Logging of FinOps and technical debt.
+6. **E**volution: Post-mortem-driven optimization.
+
+### 🎨 Pillar VI: Visual Perfection (Genesis DS)
+Design is the interface of intelligence.
+- **HSL Authority:** Colors must be strictly HSL-tokenized for consistency and accessibility.
+- **Micro-Interaction DNA:** Every component must feel alive through subtle animations and feedback.
+
+### 🛠 Pillar V: Tooling Authority
+Optimized Modern Stack:
+- **Context-Defined Architecture (e.g., React 19, TS, Vite, Tailwind v4) or per CLI.**
+- **Genesis CLI:** The orchestrator for context and boilerplate generation.
+
+### 🛡 Pillar VI: FinOps & Telemetry Rigor
+Development cost and AI resources must be precisely tracked.
+- **Token Precision:** Fact-based logging of token consumption.
+- **Saturation Monitoring:** Active tracking of model context window usage.
+
+### 🎭 Pillar VII: UI Adaptability (UI Tiers)
+The framework scales according to the project's visual needs:
+- **FULL Mode:** Complete Genesis Design System v1.1.0 experience.
+- **MIX Mode:** Adaptive tokens with external component support.
+- **CORE Mode:** pure context and modular architecture without mandatory UI dependencies.
+
+---
+*Genesis Grid AI Labs - Engineering Excellence. Visual Perfection. Zero Defects.*

package/dist/.context/eng/docs/OPERATIONS_LOG.md

@@ -0,0 +1,31 @@
+# 📓 Genesis Grid Operations Log (v1.1.0)
+
+This file is the official ledger for all autonomous and semi-autonomous operations. All agents **must** log their cycles here to ensure transparency, accountability, and context sovereignty.
+
+## 📊 Summary of Operations (FinOps v1.1.0)
+
+| Metric | Running Total |
+| :--- | :--- |
+| **Execution Time (IA)** | 00h 00m |
+| **Project Cumulative Cost** | $0.00 |
+| **Context Saturation (Current)** | 0% |
+| **Tasks Completed** | 0 |
+| **Protocol Version** | v1.1.0 |
+
+---
+
+## 📅 Chronological Activity Log
+
+| Start (ISO) | End (ISO) | Agent (Persona) | Action / Decision | Duration | Cost | Saturation |
+| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
+| 2026-01-24T00:00:00Z | 2026-01-24T00:00:02Z | system | Project initialized / updated to v1.1.0 | 2s | $0.00 | 2% |
+
+---
+
+## 💡 FinOps & Context Insights (Pillar VI)
+- **Temporal Rigor:** All Start/End fields must use the full ISO 8601 format for realistic auditing.
+- **Instance-Based Saturation:** Tracks the context window usage for the current active chat instance.
+- **Project Cumulative Totals:** The global sum of resources consumed across all development branches.
+
+---
+*Genesis Grid AI Labs - Engineering Excellence. Visual Perfection. Zero Defects.*

package/dist/.context/eng/docs/PROJECT_MAP.md

@@ -0,0 +1,35 @@
+# 🗺 Genesis Grid Project Map (v1.1.0)
+
+This document is the **Codebase Waze**. It provides the Agent with a high-level view of the project without needing to read every file, saving thousands of tokens and maintaining "Context Sovereignty".
+
+## 🏛 Domain Architecture (High-Level)
+| Module | Responsibility | Status |
+| :--- | :--- | :--- |
+| `src/lib/services` | Core services and telemetry (context.service) | [ ] Pending |
+| `src/lib/store` | Global state management (GenesisStore) | [ ] Pending |
+| `src/components/ui` | Genesis Design System Primitives | [ ] v1.1.0 Boilerplate |
+
+---
+
+## 🔗 Technical Stack (Detected Stack)
+- **Framework:** [Detected: React 19+ / Next.js / Other]
+- **Styling:** [Detected: Tailwind v4 / Genesis Tokens (HSL)]
+- **State:** [Detected: Zustand / GenesisStore / Other]
+- **Context:** Genesis Grid AI Framework (v1.1.0)
+
+---
+
+## 🛠 Entry Points & Bridges
+- **Main:** `src/main.tsx`
+- **Global Bridge:** `src/lib/bridge.ts`
+- **Telemetric Layer:** `src/lib/services/context.service.ts`
+
+---
+
+## 🚦 Observability & Health
+- **Current Context Saturation:** [Low / Mid / High]
+- **Last Context Audit:** 2026-01-24
+- **Technical Debt Log:** [None]
+
+---
+*Genesis Grid AI Labs - Engineering Excellence. Visual Perfection. Zero Defects.*

package/dist/.context/eng/docs/standards/typescript.md

@@ -0,0 +1,12 @@
+# 💻 TypeScript Standards (ENG)
+
+## 🛡 Strict Type Safety
+- **No Any:** Use of `any` is strictly prohibited. Use `unknown` or specific interfaces.
+- **Strict Mode:** Always enable `strict` in `tsconfig.json`.
+
+## 🏗 Architecture
+- **Functional First:** Prefer functional components and pure functions.
+- **Modules:** Follow the "Zenith" architecture guidelines for modularity.
+
+---
+*Engineering Excellence.*

package/dist/.context/eng/skills/agent_logging/SKILL.md

@@ -0,0 +1,22 @@
+---
+name: zenith-agent_logging
+description: Technical progress and FinOps metrics logging. Use to document every work turn in OPERATIONS_LOG.md.
+---
+---
+name: agent-logging
+description: Essential skill for logging agent operations, execution time, and FinOps costs into the PROJECTS_LOG.md.
+---
+
+# 📓 Agent Logging Skill
+
+## 📋 Protocol
+After completing a task or a significant step, the agent must update `.context/docs/OPERATIONS_LOG.md`.
+
+## 📈 Entry Format
+Add a new row to the table:
+`| DATE | AGENT_NAME | ACTION_DESCRIPTION | DURATION | ESTIMATED_COST |`
+
+Update the running totals in the summary table at the top of the file.
+
+---
+*Transparency is the foundation of scale.*

package/dist/.context/eng/skills/brainstorming/SKILL.md

@@ -0,0 +1,51 @@
+---
+name: zenith:brainstorming
+description: Creative sessions for architecture or interface. Use to explore solutions before creating an implementation plan.
+---
+---
+name: zenith:brainstorming
+description: |
+  Socratic design refinement - transforms rough ideas into validated designs through
+  structured questioning, alternative exploration, and incremental validation.
+
+trigger: |
+  - New feature or product idea (requirements unclear)
+  - Need to plan, design, or architect something new
+  - Multiple approaches seem possible
+
+skip_when: |
+  - Design already complete and validated → use zenith:writing_plans
+  - Detailed plan ready to execute → use zenith:planned_execution
+
+sequence:
+  before: [zenith:writing_plans, zenith:using_git_worktrees]
+
+related:
+  similar: [zenith:writing_plans]
+---
+
+# 💡 Brainstorming: From Ideas to Designs
+
+## Overview
+
+Transform rough ideas into well-formed designs through structured questioning and alternative exploration.
+
+**Core Principle:** Research first, ask targeted questions to fill gaps, explore alternatives, and present the design incrementally for validation.
+
+## The Process
+
+1.  **Autonomous Reconnaissance**: Inspect the repository, docs, and commits to form an initial model. Share your findings.
+2.  **Understanding**: Ask only for the missing context. Define purpose, constraints, and success criteria.
+3.  **Exploration**: Propose 2-3 different approaches with pros and cons.
+4.  **Design Presentation**: Present the design in sections validated by the user.
+5.  **Documentation**: Write the design document in `docs/plans/YYYY-MM-DD-<feature>-design.md`.
+
+## Required Patterns
+
+This skill uses these universal patterns:
+- **State Tracking:** See [skills/patterns/state_tracking/SKILL.md](file:///.context/eng/skills/patterns/state_tracking/SKILL.md)
+- **Failure Recovery:** See [skills/patterns/failure_recovery/SKILL.md](file:///.context/eng/skills/patterns/failure_recovery/SKILL.md)
+- **Exit Criteria:** See [skills/patterns/exit_criteria/SKILL.md](file:///.context/eng/skills/patterns/exit_criteria/SKILL.md)
+
+---
+*Genesis Grid - Structured Creativity.*

package/dist/.context/eng/skills/documentation_templates/SKILL.md

@@ -0,0 +1,73 @@
+---
+name: documentation_templates
+description: Documentation templates and structure guidelines. README, API docs, code comments, and AI-optimized documentation.
+allowed-tools: Read, Glob, Grep
+---
+
+# 📑 Documentation Templates (Zenith Sovereignty)
+
+> Guidelines for maintaining clarity and context sovereignty through well-structured documents.
+
+---
+
+## 1. README Structure (Golden Standard)
+
+The priority order for a Zenith README:
+
+| Section | Purpose |
+|---------|---------|
+| **Title + One-liner** | What is this project? |
+| **Quick Start** | How to run in <5 minutes. |
+| **Features** | What the system actually does. |
+| **Configuration** | Environment variables and options. |
+| **Architecture** | Link to `PROJECT_MAP.md` and technical decisions. |
+
+---
+
+## 2. Code Comments (Guidelines)
+
+### When to Comment?
+
+| ✅ Comment | ❌ Don't Comment |
+|-----------|-----------------|
+| The **WHY** (business logic). | The **WHAT** (when the code is obvious). |
+| Complex algorithms. | Every line of code. |
+| Non-obvious behavior. | Trivial implementation details. |
+| API contracts and JSDoc. | Self-explanatory code. |
+
+---
+
+## 3. Architecture Decision Record (ADR)
+
+Use this format to record critical structural changes:
+
+```markdown
+# ADR-[Number]: [Decision Title]
+
+## Status
+Accepted / Deprecated / Superseded
+
+## Context
+Why are we making this decision? What was the problem?
+
+## Decision
+What did we actually decide? What was the chosen path?
+
+## Consequences
+What are the trade-offs? What did we gain and what did we lose?
+```
+
+---
+
+## 4. AI-Optimized Documentation (llms.txt)
+
+For AI crawlers and agents, maintain an overview file:
+
+- **Clear Hierarchy**: Use H1-H3 logically.
+- **Data Examples**: JSON/YAML for data structures.
+- **Mermaid Diagrams**: For logic flows and states.
+- **Self-contained Sections**: Avoid infinite cross-references that degrade the context window.
+
+---
+
+> **Remember:** Outdated documentation is worse than no documentation. Maintain sync with code in every commit (Pillar VII).

package/dist/.context/eng/skills/exploring_codebase/SKILL.md

@@ -0,0 +1,52 @@
+---
+name: zenith:exploring_codebase
+description: Systematic discovery of project topography. Use when entering a new repository or unfamiliar module.
+---
+---
+name: zenith:exploring_codebase
+description: |
+  Autonomous two-phase codebase exploration - first discovers natural perspectives
+  (layers, components, boundaries), then dispatches adaptive deep-dive explorers.
+
+trigger: |
+  - Need to understand how a feature/system works across the codebase
+  - Starting work on an unfamiliar codebase or component
+  - Planning changes that span multiple layers/components
+
+sequence:
+  after: [brainstorming]
+  before: [writing_plans, planned_execution]
+
+related:
+  similar: [systematic_debugging]
+---
+
+# 🔍 Exploring Codebase
+
+## Overview
+
+Traditional exploration assumes structure upfront. This skill takes an autonomous two-phase approach: **discover** the natural perspectives of the codebase first, then **deep dive** into each discovered perspective with targeted explorers.
+
+## The Two-Phase Flow
+
+### Phase 1: Discovery Pass (Meta-Exploration)
+**Goal:** Understand "What IS this codebase?"
+Identify:
+- Architecture pattern (Modular Monolith, Hexagonal, etc.)
+- Major components and modules
+- Natural boundaries and layers (Bridge Pattern)
+- Core technologies and frameworks
+
+### Phase 2: Deep Dive Pass (Adaptive Exploration)
+**Goal:** Understand "How does [target] work in each discovered area?"
+- Dispatch specialists (e.g., `backend_typescript`, `frontend_specialist`) to explore their respective domains within the discovered structure.
+
+## Required Patterns
+
+This skill uses these universal patterns:
+- **State Tracking:** See [skills/patterns/state_tracking/SKILL.md](file:///.context/eng/skills/patterns/state_tracking/SKILL.md)
+- **Failure Recovery:** See [skills/patterns/failure_recovery/SKILL.md](file:///.context/eng/skills/patterns/failure_recovery/SKILL.md)
+- **Exit Criteria:** See [skills/patterns/exit_criteria/SKILL.md](file:///.context/eng/skills/patterns/exit_criteria/SKILL.md)
+
+---
+*Genesis Grid - Sovereignty through Knowledge.*

package/dist/.context/eng/skills/mcp_builder/SKILL.md

@@ -0,0 +1,64 @@
+---
+name: mcp_builder
+description: MCP (Model Context Protocol) server building principles. Tool design, resource patterns, and interoperability best practices.
+allowed-tools: Read, Write, Edit, Glob, Grep
+---
+
+# 🔌 MCP Builder (Zenith Connectivity Standard)
+
+> Principles for building robust MCP servers that extend AI capabilities.
+
+---
+
+## 1. MCP Overview
+
+The **Model Context Protocol** is the standard for connecting AI systems with external tools and data sources securely and in a structured way.
+
+### Core Concepts
+
+| Concept | Purpose |
+|---------|---------|
+| **Tools** | Functions that the AI can invoke (e.g., execute code). |
+| **Resources** | Data that the AI can read (e.g., logs, documentation). |
+| **Prompts** | Predefined prompt templates for common tasks. |
+
+---
+
+## 2. Tool Design Principles
+
+A well-designed tool is the secret to flawless orchestration.
+
+- **Scoped Name**: Use action-oriented names (e.g., `get_system_health`, `deploy_module`).
+- **Single Purpose**: Each tool should do one thing and do it well.
+- **Validated Input**: Use strict schemas with clear types and descriptions for each parameter.
+- **Structured Output**: Always return a predictable JSON format.
+
+---
+
+## 3. Resource Patterns
+
+- **Static Resources**: Config files, manuals, and fixed documentation.
+- **Dynamic Resources**: Generated on demand (e.g., real-time database status).
+- **URI Patterns**: Use clear URIs, e.g., `zenith://project/map` or `logs://server/error`.
+
+---
+
+## 4. Security and Data Sovereignty
+
+- **Full Validation**: Never trust tool input without sanitization.
+- **Secrets**: Never log API keys or secrets. Use environment variables.
+- **Permissions**: Apply the principle of least privilege for filesystem access.
+
+---
+
+## 5. Zenith Best Practices Checklist
+
+- [ ] Clear and verbal tool names.
+- [ ] Detailed descriptions (the AI needs to understand "why" a tool is used).
+- [ ] Structured error handling (not just "error," but "why it failed").
+- [ ] Logging for debugging without exposing sensitive data.
+- [ ] Well-defined mandatory input schemas.
+
+---
+
+> **Remember:** MCP tools should be simple, focused, and well-documented. The AI relies on descriptions to use them correctly in the orchestration flow.