Skip to content
Omnipresence

Agent and MCP

Connect Claude, Codex or any MCP client to your project.

Updated October 20, 2026

Omnipresence is built to be read by an agent, not just by a person. Connect your assistant over MCP and it works from your live position instead of whatever you pasted into the chat.

What the connection is

One account-scoped connection covers every project you have. Your agent lists them and switches between them, so you authorise once rather than per site. The same connection also serves the methodology itself: the processes and skills to follow, not only the numbers.

Read-only against two-way

Every plan can read. Two-way adds the fourteen write tools: logging an optimization, flagging a query as being worked, registering a brand, adding a prompt, storing a memory. The difference matters more than it sounds, because an agent that cannot record what it did starts from zero every time you open a new chat.

Context, memory and files

Three places to put information, and they are not interchangeable. Context is what the brand is: positioning, tone, audience, proof. Memory is what was decided or done, filed under one of seven facets. Files hold whole documents, up to 200 KB each, addressed by key. All three are scoped to the project, so one client's context never reaches another's.

Concepts rather than clicks: putting agents to work on this.

In this section

  1. 01Connect your agentConnect Claude, ChatGPT, Cursor or Claude Code to Omnipresence over MCP, how the account-scoped connection works, and how to rotate a credential.
  2. 02What to ask your agentNine jobs members do through the Omnipresence MCP, with the prompt to type, the tools the agent calls in order, and what comes back.
  3. 03Read-only and two-wayWhat changes when the MCP connection can write as well as read, which fourteen tools are writes, and why a loop that records itself gets better.
  4. 04Tool referenceEvery tool the Omnipresence MCP exposes, grouped by job, with its parameters and whether it needs a two-way connection.
  5. 05Brand and voice contextThe project context an agent reads before it writes anything: positioning, tone, audience, proof and keyword targets, and how it differs from memory.
  6. 06Project memoryThe notes your agent keeps between sessions, the seven facets memory is filed under, what belongs in it, and how to keep it worth reading.
  7. 07Project filesStore long-form project context an agent can load later: a brand kit, a rules registry, templates or a topic queue, keyed, versioned and listable.
  8. 08The methodologyThe MCP serves the AI SEO method as well as your data: how an agent looks up a process, fetches it, and scores a passage before you publish it.

Not tracking yet?

Seven days free, on your own model key. Point it at one brand and read what comes back.

Start your free trial