SERP Gap Analysis

Evaluate whether a keyword is realistically worth pursuing, then turn the live SERP into a compact brief that explains what content is required to compete and where the opportunity exists.

The skill has two required phases. An optional Phase 3 runs only if the user explicitly opts in after Phase 2. Each phase has a corresponding output template in templates/.

Phase 0: Collect Input and Resolve Keyword Library

Resolve the keyword library, then fetch fresh SERP data for the keyword.

Read references/input-and-library.md for the full process.

Result: a saved keyword record with top-10 SERP data and keyword metrics ready for Phase 1.

Phase 1: Quick Winnability Assessment (subagent)

Determine whether the SERP is worth pursuing before investing in deeper analysis.

Read references/winnability-rubric.md for the verdict criteria, signal definitions, and scoring rules.

Use the keyword record from Phase 0 — the saved top-10 SERP titles, snippets, URLs, positions, plus keyword-level metrics (search volume, keyword difficulty, search intent, competition level).

Deliverables:

• Winnability verdict: Winnable, Challenging, or Hard-to-Win
• Search intent brief: 1 sentence
• Key signals summary: 3-5 bullets
• Recommendation: continue to Phase 2 or skip with reasoning

Output: use template from templates/phase-1-winnability.md.

Phase 2: Deep Content Gap Analysis (subagent)

Analyze competitor weaknesses and define content requirements from the top 10 ranking pages.

Read references/content-gap-process.md for the async workflow and analysis approach.

Workflow summary:

1. analyze_serp(kw, keyword_library, mode: "content_gap_analysis") to start async analysis
2. Poll get_research_status(task_id) until complete
3. Present the consolidated result — analyze_serp already handles the full analysis internally

Deliverables:

• Must-have topics (table stakes)
• Target length range and content format recommendation
• Weakness analysis

Output: use template from templates/phase-2-content-gap.md.

After presenting Phase 2, end the reply with one natural question that does both:

• asks whether to continue and create a content brief with an outline of H2 - description sections
• asks what business, product, feature set, firsthand experience, or proprietary data the user can bring that competitors do not cover

Continue to Phase 3 only when the user explicitly says yes. Use the user's reply as the primary source of business or product context. End the skill after Phase 2 when the user declines.

Phase 3: Content Brief with Information Gain and Outline

Only runs if the user explicitly says yes after Phase 2.

Read references/information-gain.md for how to collect and apply the user's unique context.

Use the context the user provides in response to the Phase 2 question before generating the brief. Build differentiation from the user's actual context. If they say yes but do not provide enough detail, ask one short follow-up about the missing business, product, feature, experience, or proprietary-data context.

Deliverables:

• Information gain analysis
• Angle recommendations
• Outline as H2 - description pairs

Output: use template from templates/phase-3-content-brief.md.