Agent Prompt & Tool Spec Designer
Designs a complete system prompt and tool specification for an LLM agent from a description of what the agent should do.
Agent Prompt & Tool Spec Designer is a free AI skill for agent systems & llm workflows. Designs a complete system prompt and tool specification for an LLM agent from a description of what the agent should do. It works with Claude, ChatGPT, Cursor and is ready to use out of the box.
About this skill
Agent Prompt & Tool Spec Designer takes a description of an agent's intended job — what it should do, what tools it needs, and what it should never do — and produces a complete, structured system prompt plus a tool specification (name, description, parameters) for each tool the agent needs. It applies prompt engineering best practices like clear role definition, explicit boundaries, and step-by-step reasoning guidance, formatted so it can be dropped directly into an agent framework or API call. Built for developers building LLM agents who want a solid first draft instead of starting from a blank system prompt.
What it does
Given a description of an agent's purpose, available capabilities, and constraints, the skill outputs a complete system prompt (role, instructions, boundaries, output format) and a tool specification block (name, description, parameters with types) for each tool the agent would need to call, ready to use with the Anthropic API or similar agent frameworks.
What is included
- Structured system prompt template (role, instructions, boundaries, output format)
- Tool specification generator with typed parameters
- Explicit constraint/guardrail drafting
- Guidance on step-by-step reasoning instructions for the agent
- Format compatible with Anthropic API and common agent frameworks
How to use it
1. Download the SKILL.md file 2. Upload it into a new Claude conversation 3. Describe the agent's purpose, needed capabilities, and any constraints 4. Receive a complete system prompt and tool specification block
Examples
I'm building an agent that answers customer questions about order status by looking up orders in our database. It should never make changes to orders, only look things up.
SYSTEM PROMPT You are an order status assistant. Your only job is to answer customer questions about the status of their existing orders by looking up order information. You must never modify, cancel, or create orders under any circumstances — you are read-only. If a customer asks to make a change, tell them you can look up information but cannot make changes, and direct them to support@company.com for changes. When a customer asks about an order, look it up using the lookup_order tool before answering. Never guess or assume an order's status without checking. TOOL SPECIFICATION Name: lookup_order Description: Retrieves the current status, items, and shipping info for a given order ID. Parameters: - order_id (string, required): The order ID to look up - customer_email (string, optional): Used to verify the requester owns the order
FAQ
What is this skill for?
How do I install it?
Is the output compatible with the Anthropic API tool use format?
What if I don't know exactly what tools my agent needs yet?
How is this different from just asking Claude to 'write a system prompt'?
Does it write the actual code that implements the tools?
Related Skills
AI Agent Prompt & Tool Spec Builder
Designs a complete AI agent specification — system prompt, tool definitions, and decision boundaries — from a description of the task you want the agent to handle.
AI Workflow Architect
Design reliable AI-agent and LLM workflows from a business goal, including steps, tools, prompts, controls, and implementation guidance.
Prompt Engineering Coach
Turns a rough task description into a precise, high-performing prompt, then explains why each change improves reliability.
Related Prompts
Tool Definition Writer for Agents
Writes a clear, well-scoped tool definition (name, description, parameters, and usage guidance) for an AI agent, designed to minimize misuse and ambiguous tool-calling.
Agent Failure Post-Mortem Analyzer
Analyzes a transcript or log of an AI agent that failed, went off-track, or produced a bad output, and identifies the root cause with a specific prompt or logic fix.
Structured Output Format Enforcer
Designs a prompt that reliably gets an AI model to return output in a specific structured format (JSON, table, or custom schema) without malformed or inconsistent results.
Related Articles
What is an AI agent in simple words?
An AI agent is a system that uses a model to reason through a task and may use tools, memory, or workflows to complete steps beyond a single chat response.
What is prompt injection, and why should I care?
Prompt injection is when malicious or untrusted text tries to override instructions or manipulate an AI system, especially when the AI can use tools or read external content.