gooseworks-ai/goose-skills / skills/marketing/improve-search-visibility
improve-search-visibility
VerifiedSEO and AI-answer visibility. Use for a site SEO or technical audit with ranked fixes, competitor keyword and topic gaps, a topic cluster, SEO metrics for domains, a programmatic SEO page-set plan, or checking where ChatGPT and Google name the brand. Delivers the requested audit, comparison, map or blueprint with dated evidence and clear data limits. Also covers app store listings and local search. Drafts changes; never edits live. Full pages go to make-content; paid search goes to run-paid-campaigns.
references/free-pass.md
View on GitHub# The free pass
Load only for the AI-answer visibility branch. Everything here costs no credits. Do all of it before the first message to the marketer; never stop after one source to ask "go deeper?". Work through the host's own tools: browser, web search, fetch or shell — reading public pages that way is allowed (marketing-guide §3, "Public data").
## Order (cheapest and most decisive first)
0. **Place the battle** (`battlegrounds.md`): which surfaces decide findability for this business; the pre-launch path if nothing is live.
1. **The brand's own side.** Brain summary, products, approved claims status, past search runs; the homepage, pricing, about and the pages needed to cover the buying questions and relevant templates (title, first paragraph, what it says the product is); robots.txt and sitemap.
2. **Draft the question set** (the `improve-search-visibility-question-set` skill (`../../improve-search-visibility-question-set/SKILL.md`), `question-list.md`): 12–15 buyer questions — category, problem, comparison and brand, grouped into clusters — from what you just read plus customer words and the rivals' pages, covering each audience the site currently sells to. No approval needed to run it; the marketer edits it afterwards.
2b. **Read the positioning the site sells today** (homepage, product pages, blog mix, the brain) — if more than one buyer or positioning shows (e.g. old outbound posts next to new ads pages), note each for the "which buyer" decision; never pick one.
3. **ChatGPT, signed out, in the host browser** (the `improve-search-visibility-citation-check` skill, `../../improve-search-visibility-citation-check/SKILL.md`) — every question once, waiting for each answer to finish, key questions three times (`citation-protocol.md` for the settings and the wait rule). This is the headline of the first message.
4. **Google AI Overview and AI Mode in the same browser**, when Google lets the browser in. If navigation is refused or a challenge page shows, stop that surface and mark it "not checked — Google blocked the browser"; never solve a challenge.
5. **Web search for each category question** (the host's search tool): which lists, threads, videos and vendor pages come up. Label it "web search" — it is not Google's exact page one (that's the paid top-up).
6. **News and mentions, web search:** the brand and each main rival — launches, funding, press, reviews, "vs" pages, listicles that name them, in the last 12 months.
7. **Profiles answer engines read** (the `improve-search-visibility-entity-footprint` skill (`../../improve-search-visibility-entity-footprint/SKILL.md`), `entity-and-local.md`): the five that matter for this business type, read like a person would; never G2, Capterra, Trustpilot or Amazon by script.
8. **App store**, only if there is an app: Apple's public search lookup, then the listing (the `improve-search-visibility-store-listing-audit` skill (`../../improve-search-visibility-store-listing-audit/SKILL.md`), `app-store-listing.md`).
9. **Foundation** (the `improve-search-visibility-page-and-foundation-check` skill (`../../improve-search-visibility-page-and-foundation-check/SKILL.md`), `foundation-check.md`): the pages behind the gaps, crawler access by bot, CDN blocking (needs a shell), duplicate domains, structured data on the checked pages, which analytics and ad tags the site runs, and any large templated page set in the sitemap (dozens of pages on one pattern, e.g. `/brand/<name>`) — sample it before anything else: thin templated pages are a site-wide quality risk.
9b. **Rival content map** (`rival-content-map.md`): the sitemaps of the brand and 3–5 rivals, sorted by page type; templated sets sampled and scored; the page-type gap table. Reuses the brand's sitemap from step 1. Gap types that answer questions in the set become needed rows in the roadmap.
10. **Demand evidence** (`page-map.md` Demand): Google Trends and autocomplete for each cluster in the browser, Search Console when shared; a paid vendor's prompt-volume estimate only if the brand already has that account, labelled an estimate. None → "no demand data, ranked by judgment".
## Hosts without a browser
ChatGPT, AI Overview and AI Mode can't be asked. Do everything else, say plainly in the first message that the AI answers weren't checked, and offer the short paste set (`citation-protocol.md`) as one of the decisions.
## What the free pass must produce
- Per question: ChatGPT named yes/no (k of N), who is named instead, which sources it cited.
- The pattern in one sentence.
- The page-type gap table (what kinds of pages rivals have that we don't), or "not mapped" with the reason.
- Draft the requested number of evidence-backed fixes (default five for this recurring check), each in one plain line.
- What wasn't checked and why.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |