Agent Systems & LLM Workflows

Tool Selection and Agent Routing Architect

Design agent tool-selection and routing with capability matching, permissions, confidence, costs, fallbacks, retries, observability, and evaluation.

Last updated Jul 12, 2026
FreeClaudeChatGPT
TL;DR

Tool Selection and Agent Routing Architect is a free AI skill for agent systems & llm workflows. Design agent tool-selection and routing with capability matching, permissions, confidence, costs, fallbacks, retries, observability, and evaluation. It works with Claude, ChatGPT and is ready to use out of the box.

Download Skill.md Package

About this skill

Design agent tool-selection and routing with capability matching, permissions, confidence, costs, fallbacks, retries, observability, and evaluation. It uses a structured workflow, labels assumptions, and produces implementation-ready guidance with ownership, controls, edge cases, and validation.

What it does

The skill analyzes goals, users, evidence, workflows, dependencies, constraints, and risks; converts them into a practical operating model; and produces explicit rules, responsibilities, exceptions, controls, tests, and rollout guidance.

What is included

  • Agent context
  • Task and capability map
  • Tool registry
  • Selection rules
  • Permission and risk controls
  • Confidence and fallback
  • Cost and latency
  • Failure and retry behavior

How to use it

1. Download the tool-selection-and-agent-routing-architect-SKILL.md file
2. Upload it to your AI, operational, or project workspace
3. Provide the current process, users, goals, evidence, and constraints
4. Add ownership, approval, risk, and implementation requirements
5. Use the output for design, review, testing, or rollout

Examples

Example input
Design routing for an AI agent choosing between search, database queries, code execution, email, calendar, file retrieval, and human review.
Example output
A complete professional deliverable with context, decisions, ownership, workflows, edge cases, risks, controls, validation criteria, and an implementation roadmap.

FAQ

What is this skill for?
It creates a professional tool selection and agent routing architect deliverable.
Will it invent facts or results?
No. Missing evidence, assumptions, and unknowns are labeled clearly.
Can it improve an existing process?
Yes. It can audit the current process before redesigning it.
Does it include edge cases?
Yes. Exceptions, failures, escalation, and fallback behavior are included.
Can it assign ownership?
Yes. Roles, decision rights, and handoffs are made explicit.
How is this different from generic advice?
It produces a structured, testable, implementation-ready system.

Related Skills

Agent Systems & LLM WorkflowsFree

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.

ClaudeChatGPTCursor
#ai agents#llm workflows#prompt engineering
Agent Systems & LLM WorkflowsFree

AI Workflow Architect

Design reliable AI-agent and LLM workflows from a business goal, including steps, tools, prompts, controls, and implementation guidance.

ClaudeChatGPTCursor
#ai agents#llm workflows#automation design
Agent Systems & LLM WorkflowsFree

Prompt Engineering Coach

Turns a rough task description into a precise, high-performing prompt, then explains why each change improves reliability.

ClaudeChatGPTCursor
#prompt engineering#LLM workflows#AI literacy

Related Articles

Article · LLM Engineering

Building Production Systems with GPT-5.6 and Fable 5

A production-focused architecture guide for task contracts, routing, retrieval, tool security, state, validation, observability, cost controls, and human approval.

Jul 11, 20268 min read
Read Building Production Systems with GPT-5.6 and Fable 5