How do you use AI to run EOS? Connect your EOS data to an AI assistant like Claude, and it can create To Dos, audit Rocks, summarize L10s , and build dashboards for you. That’s what the Strety MCP does — and we hosted a workshop where three members of our team and two customers showed their real, daily workflows on camera.
This post is the full written version. Watch the workshop below, or read on for every use case we demoed.
What is the Strety MCP? An MCP (Model Context Protocol) server connects an AI assistant directly to your business data. The Strety MCP gives Claude, ChatGPT, or the MCP-ready AI assistant of your choice live access to your Rocks, Scorecards, To Dos, Issues, meetings, Docs, and V/TO. Instead of copying and pasting between apps, now, you just ask.
Orlando Garcia, the senior engineer who built Strety’s MCP, started using early MCPs during our own SOC 2 audit. He was bouncing between our security platform, our codebase, and Strety, copying context back and forth. His “We should have an MCP” thought turned into a beta, the beta turned into feature requests, and now it’s how a good chunk of our team starts their day.
We’re an Official EOS Licensee™ building EOS software that we run our own company on — so every use case below is something we (or our customers) actually do. This is our dogfood, and we ate it on a live webinar. Mukbang that teaches you something!
1. Create To Dos and Issues by voice, from anywhere Brian Dosal, our CEO and Integrator, was skeptical at first. He spends his life making Strety fast and well-designed, so creating a To Do inside Strety was already quicker than prompting an AI.
Then he found the real use case: the school pickup line. “The fastest way to create a To Do while on the road is just to speak into it,” he shared in the workshop. He hits the microphone in the Claude mobile app, talks, and a To Do or Issue appears in Strety — assigned, described, and waiting for the right meeting. His recent chat history is almost entirely “create this To Do, create this Issue.”
One tip from Brian: tell Claude to note that Claude wrote it. Otherwise your team wonders why you suddenly sound so formal.
2. Turn a Rock into a fully scoped project This one’s mine. I run marketing here as roughly a 1.5-person team, so project setup time is real money.
In the workshop, I created a Rock for our next MCP webinar, attached it to an empty project, and asked Claude: “Can you outline the project for me? Then we’ll make To Dos.” It came back with milestones — content, registration landing page, promotion, tech and ops — then created milestone To Dos in the backlog, then built out detailed To Dos for the first milestone.
A project skeleton that would have taken me hours of thinking can exist in about 30 minutes. The AI does the first-pass thinking; you come back and do the real thinking (the fun part!) on top of it.
3. Audit every Rock for a SMART description Even with an implementer, we’re guilty of leaving a session with Rocks like “launch the MCP server” and no definition of done.
Halfway through last quarter, Brian asked Claude: go through every active Rock in Strety and tell me which ones don’t have a SMART description. It found them — a good chunk — then commented on each Rock with a suggested SMART rewrite based on the check-ins and context it could see. That’s an Integrator’s hour of nagging, done in one prompt.
4. Wordsmith anything against your V/TO When our implementer asked us to refine our core focus, Brian gave Claude our V/TO and everything it knows about us, and asked for better drafts. Same trick works for meeting comments, headlines, and messaging: brain-dump by voice, let Claude shape it, post it back to Strety as a comment or message.
5. Mine your Docs for signals Our support team logs every feature request as a Doc in Strety, and adds comments when requests repeat. Instead of clicking through them, Brian asks Claude which Docs are getting the most comments and what’s trending. The answer arrives as a summary, and the feature pipeline gets prioritized from real customer signal. If you’re a heavy Doc user, you also now have all this data at your fingertips with the MCP.
6. Run a morning brief with your Scorecard and at-risk Rocks Roman, founder and Visionary at Cleango, a national commercial cleaning company, runs what he calls an agentic chief of staff. Every morning he asks for a brief: Scorecard metrics from the Strety MCP, Rocks that are off track and why, new Issues to chew on before the afternoon L10.
“In just one command I can understand everything that’s going on in our business,” Roman said in the workshop. He’s open about why it matters: he’s a Visionary with ADD, and the structure he’s never been able to force himself into, the agent holds for him instead, empowering him to be his most productive Visionary self.
7. Summarize L10s and prep the quarterly Roman’s workflow keeps going after the meeting. When an L10 closes, he asks for the summary — headlines, To Dos, Issues — pulled together from the Strety AI assistant’s meeting notes and the MCP. Before his quarterly, he asked Claude to prep the conversation: which Rocks to keep, which to kill, grounded in a quarter of check-ins and Scorecard history.
All that brain-dumping he does by voice during the week (ideas, Issues, To Dos for the leadership team) lands in Strety, and Strety is what’s on screen when the leadership team meets. Nothing slips between his brain and the Issues list.
8. Build a personalized dashboard on your EOS data Every single presenter converged on this one independently. Orlando asked Claude to build a live dashboard of his day — personal To Dos, project To Dos, calendar, all prioritized differently. I built one for my marketing day (Claude even used Strety branding without being asked). Roman has one that pulls his whole life into one screen.
9. Build an Executive Dashboard for Integrators And then there’s Shaz Khan, co-founder and Integrator at Tono Pizzeria 🍕. He used Claude to build a company-wide dashboard for his hospitality group that aggregates every platform they use. His reasoning stuck with me: “Comprehension is an act of communication. I’m trying to reduce friction in communication.” Same data, every person seeing it their own way.
He also pointed out that by documenting his org’s processes for AI context, he’s killing two birds with one stone by getting better EOS processes documented as well.
At the center of how his business runs is EOS, and he’s now able to find an aggregate truth in data across his tools that organizes information and workflows according to his company culture and values.
How to get started with the Strety MCP If you’re already running your business on Strety, the setup is quick — connect the Strety MCP to Claude or ChatGPT and start by asking about your own To Dos. That’s how everyone on the panel started, and the fancier workflows grew from there. For detailed instructions, check out the Strety MCP Help Doc .
If something you ask for isn’t supported yet, tell our support team. Feature requests from the beta group turned into shipped functionality within days — it’s fast for Orlando and his robots 🤖
And if you’re not on Strety yet, this is the part where I strongly encourage you to give it a try: everything above requires your EOS data to live somewhere an AI can reach it. Strety is the digital headquarters for teams running on EOS — L10 meetings, Rocks, Scorecards, Issues, projects, and people management in one place, with the MCP connecting all of it to AI. Try Strety free for 30 days — no credit card required.
FAQ What is the Strety MCP? The Strety MCP is a Model Context Protocol server that connects AI assistants like Claude to your Strety data — Rocks, Scorecards, To Dos, Issues, meetings, Docs, and the V/TO — so you can read and update your EOS data by asking in plain language.
What is an MCP server? MCP (Model Context Protocol) is an open standard that lets AI assistants securely connect to outside tools and data. An MCP server for a product exposes that product’s data and actions to the AI.
Can AI create EOS To Dos and Issues? Yes. Through the Strety MCP, Claude can create, update, move, and complete To Dos and Issues — including by voice from a phone.
Can AI help write SMART Rocks? Yes. Claude can review every active Rock in Strety, flag the ones without SMART descriptions, and comment with suggested rewrites based on check-ins and context.
Which AI assistants work with the Strety MCP? The workshop demos used Claude — the desktop and mobile apps and Claude Code. MCP is an open standard, so support extends as more AI tools adopt it.
Do I need to be technical to use the Strety MCP? No. Early MCPs required editing config files, but connecting one today happens inside the AI app’s settings. Our CEO’s main use case is talking to his phone in the school pickup line.
Does the Strety MCP work for Visionaries or just Integrators? Both. In the workshop, an Integrator used it for SMART Rock audits and feature triage, while a Visionary used it for morning briefs, L10 summaries, and quarterly prep.
