Skill Hub — AI-Assisted Development
Use AI tools safely and effectively for development work.
Skills for packaging repository context, designing tasks for coding agents, reviewing AI-generated code, selecting the right tool, and using MCP integrations without losing engineering control.
What this group covers
AI-assisted development skills help you delegate coding and analysis work without handing off accountability. They provide structured methods for preparing repo context, writing narrow agent tasks, reviewing agent output, choosing tools by workflow fit, and running MCP servers with clear boundaries.
These skills are cross-domain. They apply when you use Repomix, Claude Code, Codex CLI, Cursor, Copilot, Aider, CodeRabbit, MCP servers, or similar tools. Each page includes a working method, decision rules, templates, checklists, and agent instructions.
When to use this group
- You want to package a repository so an AI model can audit it or plan work.
- You are delegating a coding task to an agent and need it to stay narrow and reviewable.
- You need to review AI-authored changes before merge.
- You are choosing between Repomix, terminal agents, IDE agents, MCP, or code review tools.
- You are connecting MCP servers to an agent and need to define permission boundaries.
Skills in this group
Repository Context Packing with Repomix
Package selected repo context, analyze architecture, discover weak spots, and generate a safe implementation plan.
AI Coding Agent Task Design
Turn vague work into narrow, testable coding-agent tasks with constraints, validation commands, and stop conditions.
AI-Generated Code Review
Review AI-authored changes before merge: diff inspection, tests, security, public/private boundaries, and merge gates.
AI Tool Selection for Development
Choose between Repomix, Claude Code, Codex, Cursor, Copilot, Aider, MCP, CodeRabbit, and similar tools based on task type.
MCP Development Workflow
Use MCP and tool servers safely for development workflows: read-only vs write-capable tools, permission boundaries, and audit gates.
Recommended path through this group
- AI Tool Selection for Development — Decide which tool fits the task before opening it.
- Repository Context Packing with Repomix — Prepare focused, safe context for analysis or agent delegation.
- AI Coding Agent Task Design — Write a narrow, testable task for the chosen agent.
- AI-Generated Code Review — Inspect the result before merge.
- MCP Development Workflow — Add tools to agents only when boundaries are clear.
How this group connects to other skills
- AI-Assisted Development + AI-Assisted Analysis = Structured prompts, output review, and requirement validation for AI-generated work.
- AI-Assisted Development + Testing, QA, and Delivery Validation = Test generation, regression planning, and quality gates for AI-authored changes.
- AI-Assisted Development + SAP AMS = Agent-assisted operational memory, incident triage, and support tooling around SAP landscapes.
Status and limitations
This is a public working interpretation of AI-assisted development practices. It is not official vendor documentation. AI tools change rapidly; verify current capabilities, pricing, and data policies on official sites before making decisions.