heyEO for agents

Ask the question. Get a decision you can act on.

Connect heyEO to the agent where you already work. It turns measured site, audience and AI-answer evidence into a weekly brief, a next fix or an honest follow-up after you ship.

heyEO over MCP · scripted examples

The same evidence, inside the agent you already use.

Claudeheyeo.io weekly brief
heyEO connected

What changed this week?

Reply…

Illustrative answer. Your agent repeats heyEO's comparability and sample warnings.

The same evidence loop underneath

The agent changes. The standard of proof does not.

01

Inspect the evidence

Your agent reads the latest completed scans, recommendations and recorded changes.

02

Approve fresh evidence

If the answer needs another scan, you see the plan and approve it before anything runs.

03

Act on the result

Choose the next fix, keep a useful change, or revise when the evidence did not move.

Built so your agent can't surprise you

  • Asks you before a fresh scan spends credits.
  • Never edits or publishes your website.
  • Repeating the same request never starts a second scan.
  • Everything is recorded as you.
  • Says “no measurable change” when that is what the evidence shows.

Set it up in two minutes

Pick your agent. Tested with Claude.ai, Claude Code, ChatGPT, Cursor, and Lovable.

  1. Open Settings → Connectors.
  2. Choose Add custom connector.
  3. Paste the heyEO connector URL.
  4. Sign in with your heyEO account.
https://mcp.heyeo.io/mcp

Your agent signs in as you and sees only the projects you can open in the app. If you have more than one, it will ask which one.

Prompts to start with

All 8 prompts

What changed this week?

Claude · Weekly brief

Using heyEO, give me a five-minute weekly brief. Tell me what changed, what did not, and the one action worth taking next. Repeat every evidence warning.

Ready to use in your agent

What should I fix next?

Cursor · Choose the next fix

Using heyEO, find the highest-priority unresolved recommendation from my latest scan. Show the affected page, the evidence, and a definition of done. Ask before tracking or changing anything.

Ready to use in your agent

Did the fix improve the evidence?

Codex · Verify a shipped change

Using heyEO, record the change I just shipped. When a comparable follow-up is ready, tell me what was observed after it and whether I should keep, revise, or stop this line of work. Do not claim the change caused the result.

Ready to use in your agent

Questions

What can my agent actually decide?
It can turn your latest evidence into a weekly brief, identify the next recommendation worth considering, or compare a follow-up with your baseline. You still decide what to track, change and run.
Which projects can it see?
Only the projects you can open in the app. If you have more than one, your agent will ask which one, and can switch with select_project.
Can it start scans without me?
The connector plans a fresh scan first. Your agent must show you the plan and wait for your approval before starting it.
Does it change my site?
No. It reads results, runs scans and records what you tell it you changed. It never edits your website.
Can teammates use it?
Yes. Each person connects and signs in as themselves, and sees only what they can see in the app. Some actions, like setting the baseline, are owner only.