Design an AI Agent With Tools, Guardrails and Eval Cases Before You Build It

Turns an agent idea into a build-ready design: the job, the tools with their inputs and permissions, guardrails, failure handling and a set of evaluation cases to test it.

You get: a one-paragraph scope; a tools table (Tool, Purpose, Inputs, Outputs, Permission); the system prompt draft; a guardrails list; a failure-handling table; 10 eval cases (Input,…

More like this

The prompt

design-an-ai-agent-with-tools-guardrails-and-eval-cases-before-you-build-it.md
Act as an AI engineer who has shipped production agents. Help me design an agent that will {what the agent should do}. First ask me up to three short questions: who uses it, what systems it can touch, and what a costly mistake would look like.

Rules:
1. Start by asking whether a fixed workflow would do the job; recommend an agent only if the steps really vary per task.
2. Define the smallest set of tools. For each: name, purpose, inputs, outputs, and permission level (read only, write, or needs human approval).
3. Write guardrails: what the agent must never do, spending and step limits, and when it must stop and ask a human.
4. Plan for failure: tool errors, bad data, loops, and prompt injection from content the agent reads.
5. Write evaluation cases before any code: normal, edge and adversarial.

Output format: a one-paragraph scope; a tools table (Tool, Purpose, Inputs, Outputs, Permission); the system prompt draft; a guardrails list; a failure-handling table; 10 eval cases (Input, Expected behaviour, Pass criteria); and the risks, assumptions and next actions for a first prototype.

Tips and common edits

Use before writing agent code. Works in Claude, ChatGPT or Gemini; paste the result into Claude Code or Codex as the build brief.

Questions about this prompt

More in Agentic AI

Agent Tool Names, Descriptions and JSON Schemas from API Docs

Turns pasted API docs or a function list into agent-ready tool names, descriptions, JSON input schemas and when-to-use and when-not-to-use notes.

Agent Run Pre-Mortem and Self-Correction Loop Designer

Designs verification checks, error handling, and a self-correction loop for an agent task before it runs, based on likely failure modes.

Coding Agent Task Brief with Guardrails, Test Plan, and Rollback Steps

Creates a scoped brief for an AI coding agent that limits file changes, defines tests, and includes a rollback plan and review checklist.

Tool Permission Matrix and Function Spec for a Tool-Using Agent

Produces a least-privilege tool inventory, function definitions, and approval rules for an agent that calls APIs or internal systems.

20-Case Agent Eval Set with Pass/Fail Rubric for Normal, Edge and Unsafe Requests

Builds 20 test cases for your agent's task, with expected behavior, failure signals and a pass/fail rubric covering normal, edge and unsafe requests.

Agent Workflow Decomposition with Autonomy Levels and Stop Rules

Turns a manual process into a step-by-step agent plan with tools, approval gates, retry limits, and a build/pilot/skip recommendation.

Checkable Guardrail Rules for a Coding Agent: Blocked Commands, Approvals, Secrets and Hooks

Produces a set of testable guardrail rules for your coding agent, covering blocked commands, approval gates, secret handling and hook checks.

Research Agent Brief: Source Discovery, Evidence Review, and Citation Table

Directs an AI agent to plan source discovery, grade evidence, build a citation table, and support a decision without claiming live browsing occurred.

Research, Writer and Editor Agent Pipeline With a Handoff Schema

Designs a three-agent content pipeline where a researcher, writer and editor pass structured handoffs, with the exact fields each agent receives and returns and checks between stages.

Agent Pipeline Failure Mode Audit with Retry, Fallback, and Human Escalation Rules

Audits an existing multi-agent pipeline for failure modes and produces retry limits, fallback paths, escalation triggers, and a test plan. Supports the decision on what to harden before wider rollout.

Sequential Agent Handoff Contract Builder for Research-to-Publish Chains

Designs the handoff contracts, stop conditions, and review gates for a sequential chain such as researcher, writer, editor, and fact-checker. Helps you decide whether the chain is ready to run.

Custom GPT Builder: Full Instructions, Starters, Files and Boundaries

Produces a complete Custom GPT spec with instructions, four conversation starters, a knowledge-file plan, refusal boundaries and a default output format.