skills-composition

Community

Orchestrate multiple skills into safe workflows.

Authord-padmanabhan
Version1.0.0
Installs0

System Documentation

What problem does it solve?

It prevents messy, fragile agent workflows by standardizing how to chain multiple skills with clear contracts, optional dependencies, and user confirmation for risky phases.

Core Features & Use Cases

  • Chaining patterns: Build multi-step slash command workflows where outputs feed subsequent skills (or run in parallel when independent).
  • Scope resolution: Prefer project-scoped skills over personal ones using Cursor/Cline discovery order to avoid surprises.
  • Graceful degradation: Automatically detect sibling/optional skills and skip them when absent instead of failing the whole workflow.
  • HITL gates: Insert human-in-the-loop prompts before expensive, irreversible, or scope-expanding steps.
  • Error handling & contracts: Pass parameters explicitly and ensure one phase’s failure doesn’t corrupt earlier artifacts.
  • Decision guidance: Decide whether to create a new skill vs extend an existing one to reduce skill sprawl.

Quick Start

Ask the agent to design a slash command workflow that composes web-research-kb-refresh, technology-consolidation-analysis, and an optional consolidation-dashboard with a HITL confirmation before refreshing the knowledge base.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: skills-composition
Download link: https://github.com/d-padmanabhan/agent-engineering-handbook/archive/main.zip#skills-composition

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.