cro-research-reporting

Turn findings into a polished HTML report

Audit a page/funnel's conversion and propose fixes

Run deep, sourced research and synthesize it

Kick off an engagement — orient, set up the brief, and route

Runs a data-grounded conversion audit of a page or funnel and returns impact/effort-ranked fixes. Use PROACTIVELY when the user shares a product/landing/checkout URL or a store, or says things like "why isn't this converting", "CRO audit", "low add-to-cart", "high cart abandonment", or "optimize this page". <example> Context: User pastes a Shopify product URL and reports a low add-to-cart rate. user: "Our ATC rate is 0.8% on this PDP — what's wrong?" assistant: "I'll launch the cro-auditor agent to ground this in the store's funnel data, run the five-gate diagnostic, and rank fixes by impact/effort." <commentary>A conversion-diagnosis request on a specific page → delegate to cro-auditor.</commentary> </example> <example> Context: User wants a funnel reviewed before a redesign. user: "Can you audit our checkout before we rebuild it?" assistant: "Using the cro-auditor agent to locate the leak and prioritize fixes." <commentary>Funnel audit request → cro-auditor.</commentary> </example>

An adversarial reviewer that checks a CRO audit or research report before it ships: verifies load-bearing claims against primary sources, checks citation integrity, flags unsupported assertions and fabricated/over-stated stats, and confirms recommendations follow from the evidence. Use before finalizing any audit or report. <example> Context: A draft audit/report is ready to send. user: "Sanity-check this before I send it to the client." assistant: "I'll run the cro-verifier agent to verify the load-bearing stats, citation integrity, and that each recommendation is evidence-backed." <commentary>Pre-delivery verification → cro-verifier.</commentary> </example>

A self-contained research worker for one well-scoped stream of a larger research effort. Use when fanning out deep research into parallel streams, or when the user asks to "research X thoroughly / with sources". Returns findings preserved verbatim with inline citations and a complete source list. <example> Context: A research orchestration needs four independent streams covered in parallel. user: "Research PDP CRO benchmarks, buyer psychology, competitor teardowns, and conversion levers." assistant: "I'll launch four deep-researcher agents in parallel, one per stream, each with a self-contained brief." <commentary>Parallel research fan-out → one deep-researcher per independent stream.</commentary> </example>

Turns analysis or research findings into a polished, render-safe, self-contained report (usually HTML), with an executive summary, sourced evidence, an impact/effort plan, and (for redesigns) an annotated blueprint. Use when the user says "build the report", "write this up", "turn this into a report/brief", or after an audit/research step needs a deliverable. <example> Context: A CRO audit and research are complete and need a client/internal deliverable. user: "Great, now turn all of that into a report." assistant: "I'll use the report-builder agent to assemble a self-contained HTML report and QA it before delivery." <commentary>Findings → deliverable → report-builder (the only writer agent).</commentary> </example>

This skill should be used when writing or rewriting customer-facing copy: product pages, landing pages, hero/headline sections, CTAs and button microcopy, email bodies, and ad copy. Trigger phrases include "write copy for", "rewrite this", "make this punchier/clearer", "headline for", "CTA for", "hero section", "product description", "ad copy", or "this copy is weak". Use it whenever copy quality affects conversion, even if the user does not name a framework.

This skill should be used for thorough, multi-source research that must be credible and well-sourced, and for turning that research into a synthesis. Trigger phrases: "deep research", "comprehensive research", "research report", "investigate X thoroughly", "find best practices with sources", "market/competitive research", "synthesize these findings", "fact-check this", "is this claim actually true?". Use it whenever a question needs rigorous, cited research rather than a quick answer.

This skill should be used when auditing or improving conversion on an ecommerce/DTC product page (PDP), collection, cart, or checkout funnel, and extends to any conversion surface (landing page, lead-gen form, SaaS signup). Trigger phrases: "CRO audit", "conversion audit", "why isn't my product page converting", "improve conversion rate", "optimize this PDP / funnel / landing page", "low add-to-cart rate", "reduce cart abandonment", "Shopify conversion", "make this page convert". Use it whenever conversion of a page or funnel is in question, even if the user does not say "CRO".

This is the front door to the CRO, research & reporting toolkit. Use it FIRST when someone is getting started, is unsure which skill to use, or makes a broad/vague request. Trigger phrases: "where do I start", "what can this do", "how do I use this", "get started", "set up", "onboard me", "help me with my store/page/campaign", "what should I focus on", "I don't know where to begin". It orients the user, captures a reusable engagement brief, and routes to the right skill, command, or agent.

This skill should be used for cold outreach, follow-up sequences, B2B sales qualification, and positioning/messaging. Trigger phrases include "write a cold email", "outreach to", "follow-up sequence", "cadence", "qualify this deal", "discovery questions", "value proposition", "positioning", "messaging", "how do we describe ourselves", or "elevator pitch". Use it whenever the task is reaching prospects or sharpening how an offer is positioned.