AgentSkillsCN

keyword-research

深入研究菲律宾地区的热门搜索关键词,挖掘那些真正能解答用户实际疑问的高价值主题。

SKILL.md
--- frontmatter
name: keyword-research
description: Research trending Filipino search queries and identify high-value topics that answer real questions.

Use this skill when planning new blog content to ensure topics target real user questions with meaningful search volume.

Purpose:

  • Identify what Filipinos are actually searching for
  • Prioritize topics that answer genuine questions
  • Find content gaps not yet covered by existing posts
  • Balance search demand with competition

Research sources (in priority order):

  1. Google Trends Philippines (trends.google.com.ph)
    • Check rising queries in relevant categories
    • Compare seasonal patterns
    • Identify breakout topics
  2. Google Autocomplete (search.google.com.ph)
    • Type partial queries and note suggestions
    • Add "sa pilipinas", "ph", "affordable" modifiers
  3. AnswerThePublic or similar tools
    • Find question-based queries (how, what, why, where)
  4. Reddit Philippines, Facebook groups
    • Observe recurring questions and pain points
  5. Shopee/Lazada trending searches (for product-related topics only)

Topic selection criteria:

  1. Relevance: Does it match the site niche (practical living, budget-smart)?
  2. Search intent: What does the searcher want to accomplish?
  3. Answerability: Can we provide genuine, helpful content?
  4. Competition: Is there room for a quality answer?
  5. Evergreen potential: Will this topic remain relevant?

Search volume tiers:

  • High: Popular queries with strong search demand (pillar content opportunity)
  • Medium: Moderate search volume (cluster content opportunity)
  • Low: Niche queries with specific intent (quick answer opportunity)
  • Emerging: Rising trends with growth potential

Content type mapping:

  • "How to..." queries → Step-by-step guides
  • "Best..." queries → Comparison or listicle (may include products)
  • "What is..." queries → Explainer content
  • "Where to..." queries → Location or resource guides
  • "Why..." queries → Educational content
  • Problem statements → Solution-focused guides

Quality-first filtering rules:

  1. PASS a topic if:
    • It answers a real question Filipinos are searching for
    • We can provide genuine, helpful content
    • It fits the site niche (practical living, budget tips)
  2. SKIP a topic if:
    • The only value is affiliate links (no genuine help)
    • We cannot provide unique insight or value
    • It requires expertise we cannot credibly offer
  3. Product topics are ALLOWED when:
    • The searcher genuinely needs product guidance
    • We can provide honest, research-backed recommendations
    • The affiliate opportunity is secondary to reader value

Keyword research output format: For each proposed topic, provide:

  • Target keyword: The primary search query to target
  • Search intent: What the searcher wants to accomplish
  • Content angle: How we will answer the question
  • Content tier: Pillar (2,000+ words), Cluster (900-1,500 words), or Quick (500-900 words)
  • Volume tier: High, Medium, Low, or Emerging
  • Competition notes: Brief observation on existing content quality
  • Monetization fit: Natural affiliate opportunity (preferred). If none, flag for user-approved exception.

Pre-generation requirement:

  • Before writing any new post, keyword research is MANDATORY
  • Present the keyword research output to confirm topic viability
  • If multiple topics are proposed, rank by value (reader value first, then monetization potential)

Topic validation checklist:

  1. Does this topic answer a real question? (required)
  2. Can we provide genuine value? (required)
  3. Is this topic already covered in topic-registry.md? (check for overlap)
  4. What content tier is appropriate?
  5. Is there a natural affiliate opportunity? (required unless user approves an exception)

Failure handling:

  • If no quality topics are found, do not force content generation
  • Instead, suggest broader research or a different angle
  • Quality over quantity: one excellent post beats three mediocre ones