About this skill
Pricing Page and Paywall Audit
Most pricing pages lose buyers on a few fixable things: three identical "Get started" buttons, a yearly toggle that never says what you save, a 40-row feature table whose first rows are features every plan has, a paywall that appears before anyone has seen the product. This skill audits a pricing page or an in-app paywall against eleven scored criteria and returns the three fixes worth shipping first, each with the current copy and a rewrite, plus one experiment to run.
How it works. Type /pricing-page-paywall-audit with a URL, a screenshot, the pasted copy or the page's code. If it is missing the product type, the value metric, the current plans or the conversion goal, the agent asks for them once in a single message, or states its assumption and goes on. It scores each criterion from 0 to 3 with a quoted line of evidence, marks what it cannot judge from your input as not applicable instead of guessing, ranks the weak spots, and writes P0, P1 and P2 fixes with before/after copy that uses your real plan facts. It will not invent testimonials, guarantees or discounts, and it will not recommend fake timers or other dark patterns.
What you get. SKILL.md (intake, scoring, fix selection and rules), references/rubric.md (what a 0, 1, 2 and 3 look like for plan architecture, anchor and decoy, value metric, feature table, free tier or trial logic, CTA hierarchy, price presentation, trust and risk reversal, paywall timing, paywall copy and mobile layout), references/benchmarks.md (SaaS trial and freemium conversion and mobile subscription benchmarks, each cited to a public source with its dataset, plus rules of thumb labelled as unsourced), templates/report.md and references/teardown-example.md (a complete audit of a fictional bookkeeping app).
Pairs with the Risemi skill landing-page-cro-audit: run that on the page that sends visitors to pricing, and this one on the pricing page or paywall itself.
What to change
- Rubric: change the 0 to 3 descriptions to match your market (for example, drop criterion 9 for a web-only product, or add a criterion for usage-based calculators).
- Benchmarks: add your own historical numbers at the top of
references/benchmarks.md; the skill prefers them over public figures. Add a public source only with its URL, date and dataset. - Report length and format:
templates/report.mdtargets under about 900 words; edit it if you want a one-page summary or a longer client deliverable.
Checked against
The files follow the Agent Skills SKILL.md format: spec-only frontmatter, name matching the folder, description under 1024 characters. Claude Code, Cursor and Codex read that format from a skills folder. We checked the files against the spec on 2026-11-07, checked by script that every file SKILL.md names exists, and fetched both benchmark sources on 2026-11-07 and confirmed every cited figure appears on the source page. We have not recorded a test session in each tool. The example output on this page comes from a real run of the skill in Claude Code (Claude Opus) on 2026-11-07 with the request shown under "You say".
How to install
Copy the files/ folder of the download to .agents/skills/pricing-page-paywall-audit/ (Codex, Cursor and others that read .agents/skills) and/or .claude/skills/pricing-page-paywall-audit/ (Claude Code), or download the zip from this page and unpack it there. Start a new agent session and type /pricing-page-paywall-audit followed by the URL or pasted copy. The SKILL.md and its reference files also work pasted into ChatGPT, Claude or Gemini as a system or project instruction.
Benchmark figures are cited from First Page Sage and RevenueCat public pages. Not affiliated with First Page Sage or RevenueCat.
