# MCP Tool Design > Engineering robust, self-describing, and error-resilient tools for AI models. Use when this capability is needed. - Skill: `tomevault-io/mcp-tool-design` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/mcp-tool-design` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/mcp-tool-design/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/mcp-tool-design --- # MCP Tool Design A tool is only as good as its description. AI models depend on clear `JSON Schema` to function. ## Tool Definition - **Name**: Concise, snake_case (e.g., `calculate_risk_score`). - **Description**: Detailed explanation of *what* the tool does and *when* to use it. - **Input Schema**: Strict types and required fields using JSON Schema. ## Execution Logic - **Safety**: Validate inputs before processing. - **Feedback**: Provide verbose success messages or actionable error reports. - **Idempotency**: Ensure running the tool multiple times with the same input has predictable results. ## Best Practices - **Schema Evolution**: Maintain backward compatibility. - **Context-Awareness**: Use tools to supplement the model's knowledge, not replace its reasoning. --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/jcorpac) — claim your Tome and manage your conversions.