Frequently Asked Questions

Everything you need to know about the AI Agent Directory. Can't find what you're looking for? Get in touch.

About the Directory

What is the AI Agent Directory?

The AI Agent Directory is a comprehensive catalog of AI agents, frameworks, tools, platforms, and models. We help developers and businesses discover the right AI solutions for their needs — from autonomous coding agents to multi-agent orchestration frameworks.

Who is this directory for?

Developers building AI-powered applications, teams evaluating agent frameworks, researchers tracking the AI agent ecosystem, and businesses looking for AI tools to integrate into their workflows.

How are listings organized?

Listings are organized by type (agent, tool, framework, platform, model) and categories (Productivity, Development, Data Analysis, Customer Service, Creative, Research, Security, Education, Business). You can also filter by tags and search by keyword.

Is the directory free to browse?

Yes. Browsing, searching, and reading all listings and news articles is completely free with no account required.

Submitting a Listing

How do I submit an AI agent or tool?

Click "Submit" in the navigation bar to access the submission form. You'll need to provide a title, description, URL, category, and basic details about your tool. Submissions are reviewed before going live.

What types of listings are accepted?

We accept AI agents, frameworks, tools, platforms, and models. The listing should be a real, functional product or open-source project — not vaporware or concept demos.

How long does review take?

Most submissions are reviewed within 24-48 hours. We check for accuracy, completeness, and relevance to the AI agent ecosystem.

Featured Listings & Pricing

What does a featured listing include?

Featured listings get prominent placement on the homepage and category pages, a "Featured" badge, and priority in search results. Check our pricing page for current plans.

Is there a free tier?

Yes. Basic listings are free forever. You can submit your agent or tool, include a description, links, and tags at no cost.

How do I upgrade to a featured listing?

Visit the pricing page to see available plans. You can upgrade any existing listing from your dashboard.

AgentBench

What is AgentBench?

AgentBench is our open-source CLI tool for benchmarking AI agents. It runs agents against 25 standardized tasks across 5 categories (tool-use, reasoning, code generation, research, multi-step) and scores them on accuracy, latency, and cost. Results appear on our public leaderboard.

How do I benchmark my agent?

Install the CLI with npm install -g @agentbench/cli, implement a simple adapter function, and run agentbench run. Your agent just needs to accept a string prompt and return a string response — it works with any framework.

How is scoring determined?

Three scoring modes: exact-match (binary pass/fail), regex-match (pattern matching), and LLM-judge (an AI evaluates response quality 0-100). Each task specifies which mode to use. Overall scores are weighted averages across all categories.

Can I add custom benchmark tasks?

Yes. Tasks are defined in simple YAML files. Create your own tasks with custom prompts, scoring modes, and evaluation criteria, then pass them to the CLI with the --task-dir flag.

News & Content

Where does the news content come from?

Our news section covers the latest developments in AI agents, frameworks, and tooling. Articles are curated from industry sources and original reporting on the agent ecosystem.

Can I contribute articles or news?

We welcome guest contributions from developers and researchers working in the AI agent space. Reach out through our contact page with a pitch.

Still have questions?

We're here to help. Reach out and we'll get back to you as soon as possible.