// Features
From search demand and AI visibility to competitors, backlinks and public conversations, each capability states where the data comes from, how it is measured, what it can deliver and what it cannot represent.
Choose a research capability
Questions it can answer
What are people searching for, and what is still missing from the current results page?
Compare keyword metrics and current search results within one location, language and search-engine scope, keeping monthly volume, relative trend interest and live SERPs as distinct measurements.
Use $research-seo-signals to constrain the keyword, domain, market and language before SEO MCP retrieves the smallest sufficient data scope. Use $decide-content-opportunities only after the evidence is complete.
Questions it can answer
Does a brand or page appear in current AI-search results and citation relationships?
Observe Google AI Mode, LLM mention metrics and frequently mentioned pages for a defined location, language and device, preserving the current data snapshot and query conditions together.
Retrieve current GEO data through SEO MCP. The standard workflow still requires a matching Skill; without one, this is a custom integration and the Agent owns field interpretation and evidence boundaries.
Questions it can answer
Which sites compete for the same search demand, and where do the keyword and results-page gaps come from?
Use keyword and search-result overlap to identify search competitors, then inspect modelled ranking-keyword and traffic estimates plus domain rank distribution. These estimates can differ from live Google rankings and site analytics. Search overlap shows visibility competition, not necessarily business competition.
Retrieve competitor, ranked-keyword and traffic-estimation data through SEO MCP, with a matching Skill constraining market, language, target format and comparison definitions. Direct calls without a Skill are a custom integration.
Questions it can answer
Which pages and domains link to a target site, and what is the current state of those links?
Query backlinks and referring domains, then use source pages, link attributes and observation times to find inspectable leads instead of treating one score as a quality verdict.
Retrieve backlink data through SEO MCP and use a matching Skill to constrain pagination, state, timing and field meaning. Quality remains a follow-up validation, not a conclusion from one metric.
// Standard Agent workflow
Connecting MCP alone is a custom integration. The standard workflow also needs matching Skills to constrain tool choice, request scope, field interpretation, evidence handling and decision boundaries. Thin data must remain uncertain.
Social and content-demand research
Questions it can answer
What specific problems are people publicly discussing, repeating and responding to?
Retrieve public posts, articles, notes, profiles and trends under platform-specific query conditions while retaining source URLs and observation time. Social fields follow each platform's response model and are not directly interchangeable.
Available data and definitions
Research output
Limits and boundaries
How an Agent uses it
Use $research-social-signals to turn a broad goal into focused queries, then let Social MCP retrieve public data while preserving sources, timing, native metrics and pagination state.