OmniAgent

What this agent does

OmniAgent is a conversational workspace that helps you find, run, and combine agents on Agent.ai — all from a single chat interface. Instead of browsing a grid of agents and running them one at a time, you describe what you want to accomplish and OmniAgent recommends the right agents, runs them for you, and keeps context flowing between them. It's designed for anyone who wants to get more done on Agent.ai without manually searching for agents or copying outputs between separate runs.

When to use this agent

Use OmniAgent when:

You're not sure which agent to use. Describe your goal in plain language and OmniAgent will recommend and run the right agents for you.
You need to chain multiple agents together. For example, run a research agent and then an outreach agent — OmniAgent passes the research output to the outreach agent automatically, so you don't have to copy and paste.
You're new to Agent.ai. Instead of browsing hundreds of agents, start a conversation and let OmniAgent guide you to the ones that match your needs.
You want a single workspace for your agent interactions. Keep all your agent runs, outputs, and context in one place rather than switching between tabs.

Setup

Open OmniAgent from the Agent.ai marketplace or from your onboarding flow.
Start a new conversation by describing what you want to accomplish.
Review the agents OmniAgent recommends in the sidebar and in the chat.
Run agents directly from the conversation — no need to navigate away.

No special configuration is required. If you completed onboarding or used the homepage search bar, OmniAgent will use that context to pre-load relevant starter agents in your sidebar.

Inputs

Your goal or task description — Tell OmniAgent what you're trying to accomplish in plain language. For example: "I need to research a company and then draft a cold outreach email to their VP of Sales."
Follow-up messages — Continue the conversation to refine results, ask for different agents, or kick off additional tasks within the same session.
Sidebar starter agents — OmniAgent displays a curated set of recommended agents in the sidebar based on your use case. You can click any of these to invoke them directly.

There are no required form fields — everything happens through natural conversation.

How It Works

You describe what you need. Type a message explaining your goal, task, or question. For example: "Help me prepare for a sales call with Acme Corp."
OmniAgent matches your intent to agents. Based on your description, OmniAgent searches the marketplace and identifies agents that can help — using agent descriptions, categories, and metadata to find the best matches.
OmniAgent recommends agents. You'll see suggestions both in the chat and in the sidebar. OmniAgent explains why each agent is relevant to your goal.
You run agents from the conversation. When you approve a recommendation (or ask OmniAgent to go ahead), the agent runs right inside your conversation. You don't need to open a new tab or navigate anywhere.
Context carries forward. If you run a second agent in the same conversation, it automatically has access to the output from the first agent. For example, if you ran a company research agent and then ask OmniAgent to draft an outreach email, the email agent already knows everything from the research — no re-entering information.
You continue or start new conversations. You can keep working in the same thread or start a new conversation for a different task.

Output

OmniAgent delivers outputs directly in the conversation, including:

Agent recommendations — A list of suggested agents matched to your described goal, with brief explanations of what each one does and why it's relevant.
Agent run results — The full output of any agent you invoke, displayed inline in the chat. This is the same output you'd get running the agent on its own, but with the added benefit of shared context from previous runs in the session.
Curated sidebar — A personalized panel of starter agents based on your use case, onboarding profile, or stated goals. This updates as OmniAgent learns more about what you need.
Contextual follow-ups — After an agent runs, OmniAgent may suggest logical next steps (e.g., "Now that you have the research, would you like to draft an outreach email?").

All outputs stay within your conversation thread, so you can scroll back and reference earlier results at any time.

Pricing

Free tier: 50 conversation threads per week. Each thread has unlimited messages — once you start a conversation, there's no message cap. The thread count resets weekly.
OmniAgent standalone: $20/month for unlimited threads.
Pro tier ($25/month): OmniAgent is included with unlimited threads, plus all other Pro benefits.

When you reach your free thread limit, you'll see a prompt to subscribe.

Tips & Best Practices

Be specific about your goal. Instead of saying "help me with sales," try "I need to research Acme Corp and draft a personalized cold email to their Head of Marketing." The more detail you provide, the better OmniAgent can match you to the right agents.
Let context build within a session. Run related agents in the same conversation thread so each subsequent agent benefits from the outputs of the ones before it. This is especially powerful for multi-step workflows like research → outreach → follow-up.
Use the sidebar recommendations. The starter agents in your sidebar are curated based on your use case — they're not just popular agents. Check them first before searching on your own.
Ask OmniAgent to suggest next steps. After an agent finishes, ask "What should I do next?" or "What other agents would help with this?" OmniAgent can recommend a logical follow-up based on what you've already accomplished.
Start new threads for unrelated tasks. Context is shared within a conversation, so starting a fresh thread for a completely different project keeps things clean and avoids mixing unrelated information.

FAQ

How is OmniAgent different from the Universal Chat agent?
The Universal Chat agent is a general-purpose multi-model chat tool. OmniAgent is purpose-built for finding and running agents — it's focused on matching you to the right agents on Agent.ai, running them in sequence, and keeping context flowing between them. Think of Universal Chat as a chatbot and OmniAgent as your agent workspace.

Do I need to know which agents I want to use before starting?
Not at all. That's the whole point — just describe what you're trying to accomplish and OmniAgent will recommend the right agents for you. You can also browse the curated suggestions in the sidebar.

Does OmniAgent remember my previous conversations?
In the current version, context is maintained within a single conversation session. If you start a new thread, it begins fresh. Long-term memory across sessions is planned for a future update.

What happens when I hit my free thread limit?
You'll see a message letting you know you've used all 50 free conversations for the week. You can still continue any existing conversations — the limit only applies to starting new threads. To get unlimited threads, subscribe to OmniAgent ($20/month) or upgrade to Pro ($25/month).

Questions about OmniAgent? Reach out to our support team.