Marketing
Featured
$9.99Apple Search Ads
Manages Apple Search Ads campaigns through asa-cli, including account discovery, keywords, negatives, reports, and campaign changes.
apple-search-adsasa-cli
Capabase
Turn source data and query evidence into a staged programmatic SEO launch with page contracts, indexability rules, quality gates, and monitoring.
Publishing hundreds of templated pages is easy. Deciding which pages deserve to exist, which can be indexed, and how to contain a bad rollout is the real job.
This skill turns a programmatic SEO idea into a launch contract that engineering, content, SEO, and analytics can inspect. It separates observed evidence from assumptions, defines the data and page contracts, makes an indexability decision for each pilot page, catches duplicate intent and canonical conflicts, and sets pass and failure conditions for every release gate.
Use it for directories, integration pages, comparisons, locations, templates, or other data-driven page families. The package includes a decision rubric, human and JSON report formats, a canonical launch-plan schema, and a read-only checker for references, eligibility, duplicate URLs and queries, and blocker consistency. It does not fetch keyword data, publish pages, guarantee indexation, or predict rankings.
Input
Plan a US pilot for 240 integration pages. We have a product-owned integration catalog with status, supported actions, setup steps, last-tested date, and logo rights. Search evidence covers 35 integrations. Define which pages can be indexed, the page contract, internal links, release gates, and what happens when an integration is deprecated.
Output
An evidence register; field-level data contract; integration-page contract; a bounded pilot inventory with index, noindex, suppress, or unknown decisions; duplicate-intent and canonical checks; hub and contextual-link rules; rendered QA and pilot gates; freshness and deprecation containment; and a verdict tied to the exact approved cohort.
Provide the site and audience, conversion event, proposed query pattern, source datasets and rights, example records, keyword or search evidence, current site architecture, markets, rendering stack, and launch constraints. The skill returns an evidence register, data and page contracts, a pilot inventory with indexability decisions, staged launch gates, monitoring, blockers, and the next evidence-producing action.
A proposed query pattern and page family such as integrations, comparisons, locations, directories, templates, or profiles.
Representative records, field definitions, refresh ownership, and a lawful basis for using the data.
Current keyword, result-page, Search Console, crawl, analytics, competitor, or site-architecture evidence. Missing evidence can be recorded as a blocking check.
Optional. Required only to run the bundled read-only JSON launch-plan checker.
SKILL.md; programmatic SEO decision rubric; human and canonical JSON report formats; launch-plan JSON schema; read-only launch-plan checker; Agent Skills interface metadata
No reviews yet.
Input
A lead-gen site wants 12,000 service-and-city pages from a city list and one shared service description. No local availability, pricing, staff, proof, or first-party location data is available. The team says competitors do it, but no query export or conversion baseline was supplied. Return the safest launch decision and the smallest useful next test.
Output
A blocked indexable launch; explicit gaps in intent evidence, data rights, and page-specific value; suppression rather than copy inflation; a non-indexable prototype or research cohort if useful; named evidence owners; and a next action that tests query patterns and obtains defensible local data before generation.
Input
Return one canonical JSON plan for a 12-page accounting-software comparison pilot. The supplied dataset has verified plan prices and features, but four records are stale, two pairs target the same primary query, and legal review of competitor trademarks is pending. No page is approved to publish yet.
Output
One schema-conforming JSON object with stable evidence, field, contract, inventory, issue, gate, and metric IDs; stale and duplicate rows blocked from indexation; trademark review represented as an open critical issue; a blocked verdict whose blocker IDs match the record; and a non-publishing next action that can pass the bundled checker.
Updated Jul 13, 2026
Creator
Ddelvix79