skills.rest
velzicom
velzicomOFFICIAL
Explore all skills by this author →
GitHub

agent-builder

Guide creation, refinement, and troubleshooting of AI subagents in Claude Code.

Updated Mar 12, 2026
Search Tags:#troubleshooting#prompt engineering#ai development#agent#subagent#tool configuration
One-click install
npx skills add https://github.com/velzicom/nango-source --skill agent-builder-velzicom
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-builder
Source: https://github.com/velzicom/nango-source/tree/main/.claude/skills/agent-builder-skill
Command: npx skills add https://github.com/velzicom/nango-source --skill agent-builder-velzicom

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides expert guidance for creating, improving, and troubleshooting AI subagents, ensuring they are well-designed, effective, and safe.

Core Features & Use Cases

  • Agent Design: Learn best practices for defining agent scope, system prompts, and tool access.
  • Troubleshooting: Get help when agents aren't activating, have permission issues, or produce incorrect output.
  • Use Case: You need to create a new subagent to analyze code for performance bottlenecks. This Skill will guide you through defining its purpose, writing its system prompt, and configuring its tools.

Quick Start

Use the agent-builder skill to help create a new subagent for analyzing code performance.

Frequently Asked Questions about agent-builder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build an AI agent for code analysis?▼

To build an AI agent, define its specific scope, write a targeted system prompt, and configure its tool access. This skill guides you through designing specialized subagents for tasks like analyzing code performance bottlenecks.

What are best practices for AI agent design and prompt engineering?▼

AI agent design best practices include strictly defining the agent's scope, engineering precise system prompts, and limiting tool access to only what is necessary. This ensures the specialized AI assistant remains effective and safe.

Why is my AI subagent not activating or giving permission errors?▼

AI subagent activation failures and permission errors often stem from misconfigured tool access or overly broad system prompts. Troubleshooting involves refining the agent's scope and verifying its tool configuration permissions.

How do I fix AI agent output formatting problems?▼

AI agent output formatting problems are resolved by refining the system prompt and adjusting tool configuration. This skill provides troubleshooting steps to correct incorrect output and ensure the subagent generates the expected format.

Can I configure model selection for specialized AI assistants?▼

Yes, model selection is a core part of configuring specialized AI assistants. This skill covers agent design principles including choosing the right model and configuring tools to match the subagent's specific purpose.

Related Skills

Explore More Software Engineering Skills→
outfitter-devoutfitter-dev
28

claude-code-configuration

Manage Claude configuration files for MCP servers and environment variables.

Official
Intermediate
JNZader-VaultJNZader-Vault

subagent-templates

Provides structured templates for creating specialized AI subagents for software development tasks.

Official
Intermediate
mcauduro0mcauduro0
1

senior-prompt-engineer

Analyze and optimize LLM prompts with Python scripts.

Community
Advanced
korchasakorchasa
3

flow-skill-engineer-subagent

Create and configure AI subagents for specific tasks and IDEs.

Community
Intermediate