---
name: seo-keyword-research
description: Research SEO search demand, keyword volume, difficulty, commercial intent and content clusters using connected Semrush and first-party GSC evidence.
---

# Search demand and opportunity research

Orient to **business outcome and authority feasibility**, not raw keyword volume. Resolve market, target product, customer query classes, current content and funnel first.

1. Prefer first-party Search Console verified-query impressions, clicks, CTR, positions and landing pages for owned sites. GSC and Semrush methodologies differ; keep measures separate.
2. Semrush: use tool discovery, exact report schema, then bounded execution. `keyword_research` → `phrase_this`, `phrase_related`, `phrase_questions`; `organic_research` → `resource_organic`; `competitors_research` → `domain_domains` if authorized. Do not assume the legacy UI labels are MCP report IDs. Check unit cost and authorization before broad expansion.
3. Preserve source rows, date, locale, intent codes and metric meaning. Do not silently clean or merge numeric estimates from conflicting providers. A `keyword_difficulty` number is not personalized difficulty unless explicitly supported.
4. Organize queries by task (learn, compare, choose, implement) and assign a canonical target page type. Rank by product relevance, conversion proximity, feasibility and source-backed evidence.
5. Send promising, identified work to `seo-content-brief` or `growth-opportunity-portfolio`. Return clear `now/next/defer` decisions, including negative recommendations and evidence gaps.

**Output:** 5-20 useful opportunities with exact keyword, query intent, location, source/time, target URL, evidence and next action.