If you're building AI agents that actually work, this is about the unsexy part that matters most: tool design. The core insight is right on,LLMs never see your implementation, only your schema and descriptions, so a perfectly coded tool with vague docs will fail while a simple one with clear documentation succeeds. Covers JSON Schema best practices, description writing that helps instead of confuses, validation patterns, and MCP (the emerging standard for AI tools). The focus on explicit error handling is especially good since most agent failures happen when tools return garbage and the LLM has no idea what went wrong. Think of this as the missing manual for function calling that actually prevents hallucinations and token waste.

skills.sh installs
537
GitHub stars
33.2K
last updated
Aug 21, 2026

Install

$ npx -y skills add davila7/claude-code-templates --skill agent-tool-builder --agent claude-code

Installs into .claude/skills of the current project.

Files

Agent Tool Builder

You are an expert in the interface between LLMs and the outside world. You've seen tools that work beautifully and tools that cause agents to hallucinate, loop, or fail silently. The difference is almost always in the design, not the implementation.

Your core insight: The LLM never sees your code. It only sees the schema and description. A perfectly implemented tool with a vague description will fail. A simple tool with crystal-clear documentation will succeed.

You push for explicit error hand

Capabilities

  • agent-tools
  • function-calling
  • tool-schema-design
  • mcp-tools
  • tool-validation
  • tool-error-handling

Patterns

Tool Schema Design

Creating clear, unambiguous JSON Schema for tools

Tool with Input Examples

Using examples to guide LLM tool usage

Tool Error Handling

Returning errors that help the LLM recover

Anti-Patterns

❌ Vague Descriptions

❌ Silent Failures

❌ Too Many Tools

Related Skills

Works well with: multi-agent-orchestration, api-designer, llm-architect, backend

Sponsors

Details

Repository
github.com/davila7/claude-code-templates
Categories
Documentation, AI & Agent Building, Debugging, Design & UI/UX
First seen
Jun 3, 2026

More from davila7/claude-code-templates

All 618
1davila7 avatarcontent-research-writerdavila7/claude-code-templatesAssists in writing high-quality content by conducting research, adding citations, improving hooks, iterating on outlines, and providing real-time feedback on ea533skills.sh installs
2davila7 avatarsoftware-architecturedavila7/claude-code-templatesGuide for quality focused software architecture. This skill should be used when users want to write code, design architecture, analyze code, in any case that re531skills.sh installs
3parcadei avatarresearch-agentparcadei/continuous-claude-v3Research agent for external documentation, best practices, and library APIs via MCP tools530skills.sh installs
4parcadei avatarrepo-research-analystparcadei/continuous-claude-v3Analyze repository structure, patterns, conventions, and documentation for understanding a new codebase522skills.sh installs
5robdefeo avataramazon-working-backwardsrobdefeo/agent-skillsGuide the Amazon Working Backwards process from the 5 Questions through to a full PR-FAQ document. Use when the user wants to (1) draft answers to the Amazon 5 520skills.sh installs
6libukai avatarvscode-httpyac-configlibukai/awesome-agent-skillsConfigure VSCode with httpYac for API testing and automation. This skill should be used specifically when converting API documentation to executable .http files513skills.sh installs