AI & AGENTS · MCP
Your AI agent, with your project’s evidence.
Connect Claude, Codex, Cursor or another MCP client to the research, performance and audit tools in your project. Agents work from your data and a shared business brief, and save the report where your team can review it.
- 90 project tools
- 15 ready-made skills
- Plugins for Claude Code, Codex and Cursor
- Toolget_search_console_performance28 days of queries and pages
- Toolget_rank_tracker42 tracked keywords
- Toolget_audit_issuesP0–P1 findings
- SavedMonthly SEO reviewReport saved to the project
90 tools
research, performance, audits and reports
15 skills
ready-made SEO workflows
3 plugins
Claude Code, Codex and Cursor
Context
one shared brief per project
INSIDE THE PRODUCT
From a question to a reviewed report.
Agents use the same tools and data as the web app.
Connect
Connect your agent in minutes
Agent setup in the workspace gives you the MCP endpoint and instructions. Supported clients sign in with OAuth; headless and CI workflows use a managed API key.
- Claude, Codex and Cursor plugins
- OAuth sign-in
- API key for headless use
| Client | Connection | Plugin |
|---|---|---|
| Claude Code | OAuth | Yes |
| Codex | OAuth | Yes |
| Cursor | OAuth | Yes |
| Other MCP clients | OAuth or API key | — |
Skills
Start from ready-made SEO workflows
Skills guide your agent through common jobs: project setup, keyword research and clustering, competitor analysis, SEO audits, link prospecting, local SEO, AI visibility audits, crawler investigations and reports.
- 15 skills
- Repeatable steps
- Your data, not generic advice
- Researchkeyword-research · keyword-clusteringDemand and topic groups
- Competitorscompetitor-analysis · competitive-landscapeGaps and positioning
- Technicalseo-audit · request-intelligence-investigationFindings with evidence
- AIai-visibility-audit · ai-prompt-researchAnswers and citations
Context
Give every agent the same brief
Project context holds the business, audience, goals, competitors and key pages, so agents stop reconstructing the brief in every conversation.
- Shared business brief
- Competitors and key pages
- Research log
| Field | Value | Updated |
|---|---|---|
| Business | Phone retailer | 12 Oct |
| Goal | Non-branded growth | 20 Oct |
| Competitors | 3 domains | 24 Oct |
Reports
Keep the result, not the chat
Agents save self-contained reports in the project, following your templates, so findings live beside the evidence instead of in separate conversations.
- Saved reports
- Reusable templates
- Print or export to PDF
- MonthlyOctober SEO review28 Oct
- AuditTechnical priorities21 Oct
- AIAI visibility check14 Oct
HOW IT WORKS
Connect once, then ask.
Three steps from connecting your data to a decision you can stand behind.
Open Agent setup
Sign in to the workspace and follow the setup for your client.
Install the plugin
Add the Claude Code, Codex or Cursor plugin to get the skills.
Ask a real question
Start with a skill or a plain question about a project.
USE CASES
Built for the way you work.
How teams put it to work.
Monthly reviews in minutes
Ask your agent for each client’s review; it follows your template and saves the report to the project.
Investigate with evidence
Ask why traffic dropped and let the agent check Search Console, audits, crawlers and updates before answering.
ALSO INCLUDED
Built for agent work.
More capabilities that come with it, in the same project.
Costs before spend
Paid tools estimate their cost so agents can ask before they run.
Research log
Paid research is logged so the next session reuses it.
Browser extension
Inspect pages and results by hand when you need to.
FAQ
AI agents: common questions.
Straight answers about what it does, what it costs and where its limits are.
Which AI agents can connect?
Claude, Codex and Cursor have plugins with skills. Other MCP clients can connect with OAuth or an API key.
What can an agent do?
Use the project’s tools: Search Console and GA4 data, keyword and competitor research, rank tracking, site audits, URL inspection, AI visibility, reports and more.
Can an agent spend credits?
Paid tools can, and they estimate costs first. Review the scope before approving paid runs.
Where do agent results go?
Agents can save reports in the project, where your team can read, print or export them.
What are the limits of the data?
Agents act with your account’s permissions and the project’s data. Paid tools can incur provider charges; review the scope before running them. Agent output is generated analysis and needs review. The optional in-app Sam assistant is a beta.
How do I get access?
Platform access is reviewed. Request access with your website and the data sources you use, and the team will help you connect them.
See it with your own data.
Tell us about your website. We will review access and help you connect the right sources.