Join us at the Sydney AI Builders Meetup, Thu 3 Dec in Chippendale. Request a place →

An AI agent platform that starts from your goal

TobyBot is an agentic bot. One workspace holds your bots, documents, tools and models behind every goal you set.

New goalExample
Find every supplier contract in the shared drive that renews next quarter, list them with their dates and draft a reminder to each owner.

Result ready

Eight things to check in any AI agent platform

Eight things to ask of any platform, and where TobyBot answers each one.

What you needHow TobyBot covers it
Starting a jobA way for people who do not code to ask for work.A goal. The request is the whole brief.
Seeing the workSomething you can read while it runs.The plan as one card for each step, plus every tool call and result with the token count. A running task can be paused.
Repeating itWork that comes back every week.Save a goal as a bot that keeps its role, model, tools and knowledge. Run it again, or start it from a schedule, a webhook or a new Gmail message.
Your documentsAnswers drawn from your own files.A knowledge base you upload to. Agents search it with keyword and semantic search together.
Your toolsConnections to where the work happens.Native connectors for Google, Microsoft, Meta and LinkedIn, Telegram as a channel, and MCP servers for other systems, including internal ones.
ModelsRoom to pick and change models.Open models and API providers, including any OpenAI-compatible endpoint, or models you host through Xinference. Don't see the model you need? Talk to us about adding it.
Where it runsA deployment that suits the team.TobyBot cloud, or self-hosted on the Enterprise plan.
The team around itAdmin and sign-in.Admin and team workspaces, sign-in with Google (OIDC), run logs and traces.

AI agents for business: one workspace for four teams

Each team has its own page with the jobs it hands over first.

Cloud or self-hosted, with the models you choose

TobyBot runs on TobyBot cloud. The Enterprise plan also offers self-hosting, and comes with an implementation team and training.

On models, TobyBot works with open models and API providers, including any OpenAI-compatible endpoint. If you already run models through Xinference, agents can use those too.

  • TobyBot cloud on every plan
  • Self-hosted on Enterprise
  • Bring your own LLM key on every plan

Roll out your first bot one goal at a time

A platform earns its place on small jobs first. A sensible order for a team.

  1. Pick one job that repeats

    Choose work with clear inputs and a result that is quick to check, such as a weekly summary or a shared inbox.

  2. Give it knowledge and tools

    Upload the documents it needs and connect the accounts it works in. Other systems connect through an MCP server.

  3. Save it and set a trigger

    Save a goal as a bot. Start it on a schedule, from a webhook or on a new Gmail message, then add the next job.

What the TobyBot platform costs

TobyBot

  • Pro $59 a month, 2 users
  • Team $299 a month, 10 users
  • Enterprise Contact us. TobyBot cloud or self-hosted.

See every plan limit

On every plan

Unlimited bots, and you can bring your own LLM key. Pro and Team set the limits on users, actions and triggers, AI credits and knowledge base size. Enterprise is priced in a conversation with sales. The full plan details are on the pricing page.

AI agent platform: frequently asked questions

What is an AI agent platform?

An AI agent platform is software where AI agents do multi-step work with your tools, together with the parts around them: knowledge, connectors, models, triggers and a place to see what happened. TobyBot is one, built for business teams.

What should an enterprise AI agent platform include?

Look for admin and team workspaces, a sign-in method your team already uses, run logs and traces, a deployment option that suits you and a vendor who will help with the rollout. TobyBot has admin and team workspaces, sign-in with Google (OIDC), run logs and traces, TobyBot cloud or self-hosting on the Enterprise plan, and an implementation team and training on Enterprise.

Is TobyBot for developers or for business teams?

Business teams can run it from goals and templates. Developers add the parts that need code, such as MCP servers for internal systems and the API, chat widget and SDK for putting agents inside a product.

Can TobyBot connect to the systems we already use?

Yes. Systems beyond the native connectors, including internal ones, connect through MCP servers. Someone technical sets each one up once. Our guide to connecting your systems through MCP shows how.

How is TobyBot different from a chatbot or an automation tool?

A chatbot answers one message at a time. An automation tool follows a diagram someone drew. TobyBot takes a task, writes the plan and runs several steps with your tools, and you see each one. For side-by-side pages on named tools, see the comparisons.

Keep reading

Put your first team goal on TobyBot.

Book a demo and bring a job that repeats. We will set it up with you.