How to Deploy reverse-skill to Production: Complete 9-Step Deployment Guide

Deploy reverse-skill to production by selecting your platform, refreshing the tool index, running the appropriate bootstrap script, and initializing isolated case workspaces—enabling deterministic, auditable security workflows.

reverse-skill is a modular skill router that coordinates reverse-engineering, pentesting, CTF, and security-analysis workflows. Whether you're deploying on Windows, Linux, macOS, or Kali Linux, this guide covers the complete production deployment process based on the actual architecture and source code in the zhaoxuya520/reverse-skill repository.

Understanding the Deployment Architecture

The system implements a three-phase deployment model defined in ARCHITECTURE.md. Each phase maps directly to executable scripts and configuration files.

Phase 1: Platform Selection and Rule Loading

The router first determines which rule set applies. According to ARCHITECTURE.md lines 70-90, this selection drives your entire toolchain:

Platform	Rule File	Script Engine	Package Manager
Windows	RULES.md	PowerShell	winget, GitHub ZIP
Kali Linux	kali/RULES-kali.md	Bash	apt, pip, npm

Phase 2: Bootstrap Self-Installation

When tools are missing, the bootstrap system executes platform-specific installation. As implemented in ARCHITECTURE.md lines 107-123:

• Windows: skills/scripts/bootstrap-reverse.ps1 reads bootstrap-manifest.json
• Kali: kali/scripts/bootstrap-reverse.sh performs the same function

The bootstrap supports five install types: GitHub release ZIP, pip package, npm global, winget, or local HTTP MCP. Successful installation automatically updates skills/tool-index.md.

Phase 3: Skill Execution and Reporting

After toolchain verification, the router:

1. Reads skills/SKILL.md → skills/routing.md to locate the concrete sub-skill
2. Executes workflows like apk-reverse, pwn-chain, or CTF-Sandbox-Orchestrator
3. Invokes the built-in docs-generator to produce reports and update the field-journal

This behavior chain is documented in ARCHITECTURE.md lines 27-34 and lines 52-62 of README.md.

Step-by-Step Production Deployment

Step 1: Clone the Repository

Start with a secure, read-only clone:

git clone https://github.com/zhaoxuya520/reverse-skill.git
cd reverse-skill

Step 2: Choose Your Deployment Platform

Select the appropriate environment based on your operational requirements:

• Windows: Full PowerShell support with winget integration
• Linux/macOS: Bash-based deployment
• Kali Linux: Pre-configured with security tools via kali/RULES-kali.md

Step 3: Refresh the Tool Index

Populate the tool inventory so the router knows what's available:

Windows:

powershell -File skills/scripts/refresh-tool-index.ps1

Linux/macOS/Kali:

bash skills/scripts/refresh-tool-index.sh

# or for Kali specifically:

bash kali/scripts/refresh-tool-index.sh

These scripts auto-generate skills/tool-index.md.

Step 4: Verify Required Tools (Optional)

Inspect the generated index:

cat skills/tool-index.md

Step 5: Run Bootstrap for Missing Tools

If the index reports gaps, execute the platform-specific bootstrap:

Windows:

powershell -File skills/scripts/bootstrap-reverse.ps1

Kali:

bash kali/scripts/bootstrap-reverse.sh

The bootstrap reads bootstrap-manifest.json and selects the appropriate install method automatically.

Step 6: Initialize a Case Workspace

Create an isolated work/ directory with scope and timeline contracts:

Windows:

powershell -File skills/scripts/case-init.ps1

Linux/macOS:

bash skills/scripts/case-init.sh

This enforces the ops contracts defined in skills/ops/README.md.

Step 7: Route Tasks to Sub-Skills

Use the primary entry point for automatic routing:

# Auto-routed via skills/SKILL.md

cat skills/SKILL.md

Or manually invoke a specific sub-skill:

cd skills/apk-reverse && ./run.sh

The routing decision references skills/MASTER-ROUTING.md for fast-track paths and the full matrix in skills/routing.md.

Step 8: Generate Reports Automatically

The docs-generator runs automatically via the router:

python -m docs_generator

This produces the final report and updates the field-journal.

Step 9: Commit Knowledge Base Updates (Optional)

Preserve case experience for system evolution:

git add work/<case>/field-journal
git commit -m "Add experience for <task>"
git push

Key Configuration Files Reference

File	Purpose	Location
RULES.md / kali/RULES-kali.md	Platform-specific execution rules	Repository root / kali/
bootstrap-manifest.json	Installation source definitions	Repository root
skills/SKILL.md	Master workflow entry point	skills/
skills/routing.md	Complete task-to-skill mapping matrix	skills/
skills/MASTER-ROUTING.md	Fast-track routing ladder	skills/
skills/tool-index.md	Auto-generated tool inventory	skills/
skills/ops/	Scope, timeline, and evidence contracts	skills/ops/

Platform-Specific Considerations

Windows Production Deployment

The PowerShell bootstrap in skills/scripts/bootstrap-reverse.ps1 integrates with winget for native Windows packages. For corporate environments with restricted execution policies, you may need:

Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser

Kali Linux Deployment

Kali uses a specialized rule set and pre-installed security tools. The kali/scripts/bootstrap-reverse.sh script prioritizes apt packages before falling back to pip/npm installs.

Containerized Deployment

For isolated production environments, wrap the bootstrap in a container with persistent volume mounts for:

• work/<case>/ — Case artifacts and evidence
• skills/tool-index.md — Tool state
• field-journal — Accumulated knowledge

Summary

Deploying reverse-skill to production follows a deterministic, auditable process:

• Select platform rules — RULES.md or kali/RULES-kali.md gate execution
• Refresh tool index — Run refresh-tool-index.ps1 or .sh to detect available tools
• Bootstrap missing dependencies — Execute bootstrap-reverse.ps1 or .sh which reads bootstrap-manifest.json
• Initialize isolated workspaces — Use case-init.ps1 or .sh for scope enforcement
• Route automatically — The master SKILL.md dispatches to sub-skills via routing.md
• Generate reproducible outputs — docs-generator creates reports and updates field-journal

This architecture scales from single analyst workstations to hardened production environments without manual intervention.

Frequently Asked Questions

What is reverse-skill and when should I deploy it to production?

reverse-skill is a modular skill router for reverse-engineering, pentesting, CTF, and security-analysis workflows. Deploy it to production when you need deterministic, repeatable security operations with automatic tool management and auditable evidence chains. The system enforces scope contracts and maintains operational knowledge through the field-journal mechanism.

How does the bootstrap script know which tools to install?

The bootstrap scripts read bootstrap-manifest.json to determine the correct install type for each missing tool. According to ARCHITECTURE.md lines 107-123, supported methods include GitHub release ZIP, pip packages, npm global installs, winget, and local HTTP MCP endpoints. The bootstrap updates tool-index.md after successful installation.

Can I deploy reverse-skill without running the bootstrap?

Yes, if your environment already contains all required tools. Skip bootstrap by ensuring skills/tool-index.md shows all tools as available after running refresh-tool-index.ps1 or .sh. The router verifies tool presence before executing any sub-skill workflow, so manual pre-installation is fully supported.

What happens to case data after I complete a security task?

The case-init scripts create isolated work/<case>/ directories with embedded ops contracts. Upon completion, docs-generator produces a final report and appends experience to field-journal. You can optionally commit this journal to grow the system's knowledge base for future similar tasks, as documented in skills/ops/README.md.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}