Agent Tool Builder
- 522 installs
- 29.9k repo stars
- Updated July 27, 2026
- davila7/claude-code-templates
agent-tool-builder is a Claude Code skill that designs JSON Schema tool definitions, descriptions, validation, and error responses for developers who need Claude-style agents and MCP tools to call functions reliably with
About
agent-tool-developer is an agent skill for developers shipping AI products who need tools that work on the first call—not loops, silent failures, or mystery 10x token bills. It treats the schema and natural-language description as the real API surface, because models never execute your Python; they only choose arguments against what you documented. You get patterns for crisp JSON Schema, examples that steer usage, validation boundaries, and errors that teach the model how to retry. MCP is framed as the practical lingua franca when you want the same tool usable across Claude Code, Cursor, and other hosts. Use it while defining new capabilities in Build, and again in Ship or Operate when tightening production tool contracts after real agent traces.
- Centers design on what the LLM reads: schema and descriptions, not hidden implementation
- Covers JSON Schema patterns for unambiguous parameters and recoverable errors
- Documents input examples and description writing that reduce hallucinated calls
- Includes MCP tools as the emerging standard for agent-world interfaces
- Calls out anti-patterns: vague schemas, silent failures, and token-wasting tool loops
Agent Tool Builder by the numbers
- 522 all-time installs (skills.sh)
- +20 installs in the week ending Jun 18, 2026 (Skillselion tracking)
- Ranked #1,701 of 16,659 AI & Agent Building skills by installs in the Skillselion catalog
- Security screen: LOW risk (skills.sh audit)
- Data as of Jul 28, 2026 (Skillselion catalog sync)
npx skills add https://github.com/davila7/claude-code-templates --skill agent-tool-builderAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 522 |
|---|---|
| repo stars | ★ 29.9k |
| Security audit | 3 / 3 scanners passed |
| Last updated | July 27, 2026 |
| Repository | davila7/claude-code-templates ↗ |
How do you design reliable agent tool schemas?
Design JSON Schema tool definitions, descriptions, validation, and error responses so Claude-style agents call functions reliably—including MCP-shaped tools.
Who is it for?
Backend and AI engineers defining agent function catalogs who need schemas and descriptions that reduce hallucinated arguments, loops, and silent tool failures.
Skip if: Developers only implementing business logic inside tools without touching schema design, descriptions, or MCP registration contracts.
When should I use this skill?
A task involves designing agent tools, JSON Schema for LLM functions, MCP tool shapes, or fixing unreliable tool-calling behavior.
What you get
JSON Schema tool specs with LLM-oriented descriptions, validation rules, and standardized error response shapes ready for agent or MCP registration.
- JSON Schema tool definitions
- MCP-compatible tool manifests
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
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la462k185k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.379k15.8k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.378k92.5k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.375k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k8How it compares
Pick agent-tool-builder when the bottleneck is tool schema and description design rather than general prompt engineering alone.
FAQ
What does agent-tool-builder cover beyond implementation?
agent-tool-builder focuses on JSON Schema fields, LLM-facing descriptions, validation rules, and error response contracts so agents pick and call tools correctly, treating descriptions as critical to reliability.
Does agent-tool-builder support MCP tools?
Yes—agent-tool-builder includes MCP-shaped tool design guidance so developers can align function definitions with the emerging MCP standard used across Claude-style agent integrations.
Is Agent Tool Builder safe to install?
skills.sh reports 3 of 3 security scanners passed. Review the Security Audits panel on this page before installing in production.