AI Agent Use

THE PRACTICAL SIDE OF AI

AI agent directory.
Practical guides & comparisons.

Find agents, models and tools for your next workflow. Explore practical guides, inspect test results and try useful tools.

Explore the directoryRead the guides

Curated tools · Source-linked guides · No invented test scores

EXPLORE THE LATEST

All guides
DECISION MODEL

Laya

Classify incoming messages before your workflow takes action. Read the guide and inspect actual local results.

8-case local testExplore Laya
CLAUDE MODELS

Sonnet 5.5 / Opus 5.5

Compare published API rates, estimate token costs and download a repeatable task-testing worksheet.

Source-based comparisonCompare the models
BROWSER TOOL

Hotel Lobby Prompts

Build an editable video brief with your subjects, camera and scene. Copy or download the prompt.

Free prompt builderBuild a prompt

THE DIRECTORY

Agents & workflow components.

Browse the directory
LLocal smoke test

Laya

Route messages and classify tasks with a decision model. Explore our eight-case local smoke test.

Decision model · Agent component
MOfficial source

Muse

Meta’s personal agent for everyday tasks. Start with a small, reviewable assignment.

Personal assistant
COfficial source

Claude

An AI assistant for working through writing, analysis, and coding tasks.

Research & writing
nOfficial source

n8n

Build repeatable workflows that connect tools, data, and AI steps.

Workflow automation
BOfficial source

Browser Use

Browser automation tools and infrastructure for agent-driven workflows.

Browser automation

LESS GUESSWORK. BETTER BRIEFS.

Start with a useful workflow.

01
GETTING STARTED · 4 MIN READ

Give your agent a brief it can actually follow.

Turn a vague request into a clear task, with a definition of done.

02
TESTING · 5 MIN READ

Test the task. Not the demo.

A repeatable scorecard for accuracy, effort, and useful results.

03
USE CASE · 4 MIN READ

Build a research shortlist with traceable sources.

A practical workflow you can adapt to your next decision.

Evidence over enthusiasm.

We label vendor claims, editorial guidance, and hands-on testing separately. Read source-based guides and inspect the raw results from our first small Laya email-routing test. Each page explains what was and was not tested.

How we work

What is an AI agent directory?

An AI agent directory helps you find tools by the work you want to complete. This directory includes personal assistants, research and writing tools, workflow builders, and browser automation tools. A listing is a starting point for evaluation, not a guarantee that a tool will complete your task.

Choose an agent in three steps

  1. Browse AI agents by task and shortlist tools that fit your workflow.
  2. Write a task brief with inputs, constraints, and a checkable output.
  3. Evaluate the result using accuracy, completion, and review effort.

AI agent directory FAQ

Does a listing mean the tool was tested?

No. Directory entries use linked vendor sources. We label untested tools and do not assign performance scores without an independent test.

Where should I start with Muse AI?

Read the Muse AI guide for the official source, an availability checklist, and a first-task prompt.

Should I choose an agent or a workflow builder?

Start with how repeatable your task is. Use our AI agent vs workflow automation comparison to decide which approach to investigate.