Run on a schedule
Grok Automations run a job you describe on a schedule, or when an email arrives. In Claude Cowork, scheduled tasks on paid Claude plans run instructions you save, on a cadence you choose.
OpenTek Solutions
Consulting that gets your company on board with AI, using Grok bots and Claude, and shows your team how to use it safely and securely, with a person approving anything that leaves the building.
Independent consultancy. Not affiliated with, endorsed by, or sponsored by xAI or Anthropic.
Grok bots and Claude can take on whole jobs. The details below are what xAI and Anthropic describe in their public docs.
Grok Automations run a job you describe on a schedule, or when an email arrives. In Claude Cowork, scheduled tasks on paid Claude plans run instructions you save, on a cadence you choose.
A Grok bot can use connectors such as Gmail and Slack, and a browser on its own cloud computer. Claude supports custom connectors through remote MCP.
Grok bots run on a persistent cloud computer, so work continues after the laptop is closed. Claude scheduled tasks run in Anthropic's cloud and do not need a device to stay online.
A Grok bot can save a workflow as a skill and run it again. On Claude Team and Enterprise plans, owners can provision skills for the organization, and people can publish a skill org-wide when an owner allows it.
In Grok Bot, a group chat holds two to six bots. They can hand work to each other in one thread.
Grok can leave email and Slack as drafts you approve, and supports rules such as “Ask first before sending any external email.” Claude asks for approval before it acts in a connected app. On Team and Enterprise plans, an owner can require approval, always allow, or block each action.
We help you choose. Some companies use both. Everything runs in your own accounts.
xAI
Named assistants on a persistent cloud computer, with a browser and connectors such as Gmail and Slack. Grok Automations run a job on a schedule. A group chat can hold two to six bots. Sending stays behind drafts and rules you set, including ask-first before an external email.
Anthropic
Cowork scheduled tasks on paid Claude plans run in Anthropic's cloud. Custom connectors use remote MCP. Team and Enterprise plans can share skills across the organization. Claude asks before it acts in a connected app, and owners can tighten or allow each action.
01
An assessment of how work gets done. We pick one to three starting jobs, choose a platform, and agree what AI may and may not do.
02
Data handling, least-privilege connectors, approval rules, and an AI use policy, before anything is connected.
03
In your own account, read-only and drafts-first at the start.
04
Coaching so your people can run and extend it, plus check-ins and advisory until it sticks.
You own everything we set up. It runs in your own Grok or Claude account. We never keep your logins. Plans come directly from xAI or Anthropic, and OpenTek does not resell them.
Small businesses
Consulting for small businesses that want ChatGPT, Gemini, Perplexity, Grok, and Google AI Overviews to describe and recommend them.
We review how those assistants talk about the business today, which competitors they name instead, and gaps in the site, content, listings, reviews, and structured data. You get an assessment and a prioritized fix plan. Hands-on help is optional.
We start read-only and drafts-first. Approval rules are narrow, and we test them so they stop the actions that should not happen on their own. Each connector gets the least access that still lets the job get done.
We also help companies that follow HIPAA guidelines onboard AI.
How we handle safety and security →
Heath Kesler has decades of enterprise software behind this practice. He is an Apache Karaf committer and a co-author of Apache Karaf Cookbook. Experience contributing to technologies including Apache Karaf, Camel, and ActiveMQ.
About Heath →Our product
Reveal Me shows how AI assistants, including ChatGPT, Gemini, and Grok, find, describe, and recommend your business, and what to fix.