Skip to main content

AI assistant integration · Powered by MCP

Ask your BrandPresence evidence where you already work.

Connect an MCP-compatible AI assistant to investigate findings, compare provider runs, retrieve retained evidence, and prepare an evidence-backed worklist using the structured records already held in your workspace.

Read-only by default · scoped writes · OAuth + PKCE · revocable

Connected assistant

BrandPresence evidence workspace

Which three issues should the team address first?

The strongest current priorities are the critical crawl finding, the missing organisation schema on commercial pages, and the retained provider answer/source evidence attached to the question.

Finding · CRITICAL · 14 affected URLsAnswer Ledger · AI visibility · source linkedExperiment · awaiting after-change run

The workflow

Connect once, then investigate or act within scope.

BrandPresence exposes a bounded MCP surface. Access begins read-only; an organisation owner or admin can separately approve narrow management scopes for supported product actions.

01

Choose an AI assistant

Use an MCP-compatible client that supports remote Streamable HTTP and OAuth discovery, or a manual bearer token when required.

02

Approve organisation access

An owner or admin chooses the organisation and grants only the read or management scopes the connection needs.

03

Investigate, then review actions

Inspect retained evidence first. When a management scope is present, supported actions still pass the same role, plan, feature, usage and lifecycle checks as the UI.

Questions to ask

Start with work your team already does manually.

The integration is useful when the answer needs the retained evidence behind the dashboard rather than another generic explanation of AI visibility.
→

Which critical findings should we fix first, and what evidence supports that order?

→

Where does Perplexity mention us while OpenAI and Gemini do not?

→

Explain the evidence behind the highest-priority observed gap.

→

What changed across our latest AI monitoring runs?

→

Which experiments are still waiting for after-change evidence?

Available evidence

A structured surface across the BrandPresence workspace.

Each tool is schema-validated, paginated, organisation-scoped, and bounded to the evidence categories approved for the connection.
Sites and bounded workspace overviews
Technical findings and occurrence history
Answer Ledger entries and retained source trails
AI and technical monitoring runs
Experiments and measurement status
Search performance and usage summaries

Least privilege by design

Credentials are read-only by default. Management scopes are separately granted, role checked and audited. No scope permits autonomous customer-site editing or publishing.

Organisation-scoped access

Connections inherit an explicit organisation boundary. Supplied site and entity identifiers are filters rather than authorization.

OAuth-first connection

Compatible clients can discover authorization from the MCP endpoint. Manual scoped tokens remain available for clients that require custom headers.

Use the evidence beyond the dashboard

Build the workspace, retain the evidence, then bring it into your assistant.

AI assistant access is included with BrandPresence monitoring plans and remains under organisation-owner control.

Create your workspace