Free plan · 1,000 credits a month

Go beyond the chatbot. Do deep work with AI.

Whatever your work, describe the outcome you want in plain language. A team of AI agents plans it, uses the tools you allow and delivers a finished result. Build AI workflows for the jobs you repeat without writing code, then bring them into the apps, chat tools and systems you already use. Every agent is monitored, governed and kept within your limits.

  • Deep work, not just chat
  • Build your own AI tools
  • Works inside the apps you use
  • Monitored, governed agents
Research and report · example task Demo
Lead agentResearchAnalysisShareneeds approval
  • ResearchCompare the options40 creditsDone
  • SharingEmail the report60 creditsApproval
  • AnalysisAssess risks and trade-offs20 creditsWorking
  • SourcesGather supporting evidence10 creditsDone
Emailing the report is held back: your rules require approval. Example AI usage today: 310 of a 5,000-credit daily limit.
Works with
QuickBooksXeroMicrosoft 365Google WorkspaceShopifyHubSpotSalesforceZendeskSlackMicrosoft TeamsGoogle ChatWhatsApp BusinessNotionAirtableAnthropicOpenAIGoogle Gemini
Why Aktor

AI that does the work, not just the talking.

Most people meet AI as a chatbot: useful answers, but you still do the work. Aktor gives everyone, technical or not, a team of agents that carries the work through to a result. Build AI tools for the jobs you repeat and bring them into the apps and systems you already run. You stay in control of what those agents can do and spend.

A chatbot stops at an answer

Real work takes research, comparison and careful analysis, then a report, proposal or plan you can actually use. A chat window leaves you to do the legwork and stitch the pieces together.

AI sits apart from the work

The person who knows the work best is rarely the one who can code it. And an AI tool only helps when it reaches the apps, chat tools and older systems people already work in.

Agents need oversight

Before you hand work to AI, you need to know what your agents are doing, which tools and data they can use, and what they're spending, with limits they can't get around.

Bring your expertise. Let agents do the legwork. You define the work and judge the result. Aktor coordinates the agents, and monitors, governs and controls them along the way.

How it works

From a question to work you can use.

  1. 1

    Describe the outcome

    Ask in plain language: research a question, compare your options or draft a plan. Say what the finished work should look like. No coding or agent-building experience required.

  2. 2

    Set the boundaries

    Attach documents, spreadsheets or images, and use the business tools your admin has connected. Set a spending limit and work within your team's access and approval rules.

  3. 3

    Watch, refine and reuse

    See the agents working and pause or stop them when needed. Refine the result with follow-ups. When it's work you repeat, turn it into a workflow your team can run again.

Your expertise, in plain language

A clear brief is enough to get started.

You know the question, the context and what a useful result would look like. Give that brief to your AI team. Agents can divide the research, compare findings and assemble a report or presentation. Follow their progress, question their conclusions and ask for revisions in the same conversation.

  • Use your documents, spreadsheets and approved connections as context
  • Ask for supporting sources, trade-offs and unanswered questions
  • Download the result and keep refining it with follow-ups
Your brief · example research task
Help me decide which option to go with.

Use the documents I've attached.
Research what others have done and what it costs.
Compare the options, their risks and trade-offs.

Recommend one, with supporting sources.
Point out assumptions and gaps in the evidence.
Create a short presentation I can share.

Use only the tools I've allowed.
Keep within the AI spending limit I've set.
Your follow-up · refine the result
Compare the two strongest options in more detail.
Explain where the evidence is thin.
Update the report and presentation.
Features

Delegate the work. Keep control of it.

Delegate complex work

A lead agent plans the work and brings in specialists where useful. They research, analyze and share findings before assembling the result. You set the goal and review what they produce.

Set budgets before work starts

Preview estimated AI usage and time, then set a task budget or daily workspace limit. Track spending as agents work, and the platform stops them at the limit.

Bring your files and business tools

Give agents documents, spreadsheets and, with a vision-capable model, images to inspect. Connect tools such as Microsoft 365, Google Workspace, QuickBooks or HubSpot for more context.

Decide what agents are allowed to do

Control the tools and data agents can access. In workflows, require approval before emails are sent or business records are changed. Agents cannot grant themselves extra permissions.

Monitor live work and review each run

Watch the agent graph, messages, actions and spending as work progresses. Inspect recorded steps afterwards. Workflow actions and approvals also appear in a tamper-evident audit log.

Keep your business data separate

Your organization's records, conversations and runs are isolated from other organizations. The platform checks access when agents use tools, so their instructions cannot override your permissions.

Build your own AI tools

Describe a job you repeat and Aktor drafts a workflow of agents for it. Change it in plain language, then run it on a schedule, from a form or when new information arrives. No code needed.

Bring AI into the apps you already use

Package your workflows, data and an assistant as a Solution. Put the assistant on your site or in Slack, Microsoft Teams or Google Chat, and connect existing systems through its API and webhooks.

Give each teammate the right access

Owners and admins manage connections and team rules. Roles decide who can build, run or view work. Teammates use approved tools without handling the credentials behind them.

  • Know where AI will pay off firstHow Aktor helps: A free assessment ranks which workflows to automate AI readiness
  • Start with a question or goalHow Aktor helps: Agents plan and divide the work Tasks
  • Get a useful deliverableHow Aktor helps: Reports, spreadsheets and presentations Files
  • Turn repeated work into a toolHow Aktor helps: Workflows built and changed in plain language Workspaces
  • Bring AI into existing appsHow Aktor helps: An assistant, chat channels, an API and webhooks Solutions
  • Follow the work as it happensHow Aktor helps: Live agent activity, messages and spending Agent graph
  • Control what agents can doHow Aktor helps: Tool permissions and workflow approvals Safety policy
  • Keep AI spending within limitsHow Aktor helps: Enforced task and workspace budgets Budgets
  • Review and improve the resultHow Aktor helps: Recorded steps, files and follow-up conversations Run history
Where to start

Put AI to work on the questions that matter.

Work through a big decision

Ask agents to research the options, gather evidence and weigh the trade-offs. Get a recommendation that shows its sources, its assumptions and the questions still to resolve.

Make sense of complex information

Bring together documents, spreadsheets and notes. Ask the team to find patterns, explain differences and compare options, then discuss the findings before deciding what to do.

Build a tool for work you repeat

Describe a job you do every week, such as a status report or a round-up of new requests. Aktor turns it into a workflow anyone on the team can run, with the same limits and approvals each time.

Not sure where AI will help most? Take the free AI readiness assessment: describe a few jobs your team repeats and see which ones Aktor can automate now, how much oversight each needs and what to set up first. Then start with a Task for one-off research, analysis or a proposal. When the work repeats, build it into a Workspace: your own AI tool, described in plain language and run on a schedule, from a form or when new information arrives. To put it in front of customers or connect it to your other systems, package it as a Solution.

For the people who know the work

Give your team AI they can direct. With controls you can enforce.

The people closest to the work know which questions matter and what a good result looks like. Give them a way to investigate, analyze and prepare recommendations with AI, and to build the tools they need, without becoming developers.

Owners and admins set access, connections and spending limits. Teammates work within those rules and can see what their agents are doing. Workflows add approval steps for sensitive actions, with a record you can review afterwards.

Monitored, governed, controlled
  • Choose which tools and data agents can access
  • Set task budgets and daily workspace spending limits
  • Require approval for sensitive workflow actions
  • Watch progress and pause or stop a task when needed
  • Review recorded actions, messages, results and spending
Monthly report · example workflow Demo
“Each month, gather the latest figures from our spreadsheet, explain what changed and prepare a short report. Use read-only access to Google Sheets. Ask for approval before emailing the report.”
  • RunsMonthly
  • Google Sheets accessRead-only
  • Email deliveryApproval required
  • Example AI usage today120 of 2,000 credits
In this example, the report is ready to review. Email delivery waits for approval; the underlying records remain unchanged.

Researchers & analysts

Gather sources on a question, compare what they say and prepare a summary with the evidence behind each conclusion.

Managers & operators

Compare proposals or options, weigh the costs and trade-offs, and prepare a recommendation with open questions.

Educators & trainers

Turn handbooks, notes and policies into guides, course material or an onboarding plan, ready for review.

Writers & marketers

Research a topic or audience, draw on your own material and draft a brief, article or tailored proposal.

Start with one meaningful piece of work. Give more teams the same ability to delegate, monitor and review it.

Questions, answered

Can my team use Aktor without writing code?

Yes. Describe the research, analysis or deliverable you need in plain language, attach relevant files and review the work in the dashboard. You don't need to write prompts for every agent or build an integration to start with your own files. Owners and admins manage connected tools and shared rules. Custom integrations and self-hosting may need technical support.

Where should we start with AI?

Take the free AI readiness assessment. In about five minutes, describe a few jobs your team repeats, in your own words. Aktor checks each one against four questions: can it reach the tools involved, does it need judgement or just fixed rules, are the rules written down, and what happens if it gets something wrong. You get a score and a heatmap for every workflow, a ranked list of quick wins, how much oversight each one needs, and what to put in place first. Pick one and Aktor drafts the tool for you to review. The assessment is included in every plan, Free too, and uses a small part of your monthly credits. Run it again whenever your tools or processes change.

Can I build my own AI tools without coding?

Yes. Describe a job you repeat and Aktor drafts a workflow: a team of agents with a stage for each part of the work. Change it by saying what you want, or start from a template. Run it on demand, on a schedule, from a form or when new information arrives. Each run stays within the budgets, permissions and approvals your admin has set.

Can I connect Aktor to the apps and systems we already use?

Yes, with Solutions. A Solution packages your workflows, data and an assistant as one app. Describe it in plain language or start from a template such as a helpdesk. Then put its assistant on your website or in Slack, Microsoft Teams or Google Chat. Existing applications, older ones included, can run its workflows, read and write its records and receive webhooks over a standard REST API with an OpenAPI description. Copy-ready code is provided for Next.js, Express, Python and plain HTTP. Spending limits, permissions and approvals still apply, with a spending cap for each of your customers.

How is this different from using AI as a chatbot?

A conversation is how you give the brief and refine the result. Aktor then coordinates agents that can break down the goal, use permitted tools, exchange findings and produce files across multiple steps. You can monitor their activity and spending throughout. Review important findings and supporting sources before acting on them.

Which tools can it work with?

Accounting (QuickBooks, Xero), office (Microsoft 365, Google Workspace), sales (HubSpot, Salesforce, Zoho CRM), shops (Shopify), support (Zendesk, Freshdesk, Intercom), projects (Notion, Jira, Asana, monday.com), data (Airtable, Google Sheets, BigQuery, your own database read-only), HR (BambooHR), GitHub, plus Slack, Microsoft Teams, WhatsApp Business and any REST API or MCP server.

How does pricing work?

Each plan includes a monthly allowance of AI usage, measured in credits, and sets which models, workspaces and seats you get. Free includes 1,000 credits a month. Starter is $8 a month with 5,000 credits, Pro is $15 with 8,000 credits and every model, and Team is $25 with 12,000 credits and room for 25 people. If the allowance runs out, buy more credit or upgrade. Usage on your own model API key is billed by that provider and doesn't use credits. Enterprise, for self-hosting on your own infrastructure, has custom pricing.

How do I control AI spending?

Task previews estimate AI usage and time as a range. Actual usage depends on the model and the work involved. Set task budgets and daily workspace limits, then track spending as work runs. The platform enforces those limits; the estimate is a guide, not a fixed-price quote.

Do I need my own AI model account?

You can use the models included in your plan without adding your own API key. On every plan, including Free, an admin can also connect your organization's model account or a local model. Aktor supports Anthropic, OpenAI, Google Gemini, Amazon Bedrock and Ollama. We'll help you choose a setup during onboarding.

Who can see our business data?

Your organization's records, conversations and runs are kept separate from other organizations, and roles decide who in your team can see what. On the hosted service, our operators manage organizations, plans and membership for support, and their actions are recorded. For full control over who can reach your data, self-host on the Enterprise plan. Data sent to a connected tool or model provider is handled under that service's terms.

Which actions need my approval?

Your admin sets the rules for sensitive actions, such as sending an email or changing a business record. In workflows, an action covered by an approval rule waits for a person. One-off tasks block actions that require approval rather than running them automatically. Agents can't loosen the rules or grant themselves more permissions.

Can I run Aktor on my own servers?

Yes, on the Enterprise plan. We help you deploy Aktor on your own infrastructure with Docker, and you can use local models through Ollama to keep model processing there too. External model providers and connected cloud tools still receive the data needed for the requests you send them. Talk to us at info@quabcore.com.

Is my data used to train AI models?

Aktor doesn't train AI models on your data. When a task uses an external model, the relevant inputs go to that provider under its API terms and data-use settings. Your admin chooses which models are available to your organization.

Bring the work. Your AI team is ready.

Start with a free AI readiness assessment: describe a few jobs your team repeats and see which ones AI can take on. Or bring a research question, a challenging analysis or a plan you need to develop. The Free plan includes 1,000 credits a month, with no card needed.