Skip to content
SkillModel-led395 tokens in use

ads-landing

Audit the page an ad clicks through to: message match, mobile layout, speed, trust, forms, consent and tracking, with a guard against private-network fetches.

11,191 stars on the repository MIT LicenseAgriciDaniel/claude-ads · skills/ads-landing
/plugin install claude-ads@ai-marketing-hub-claude-ads
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing ads-landing costs

81tokens idle
395tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter8/15
  • Structure15/25
  • Examples3/15
  • Output15/15
  • Size13/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity84
  • Maintenance100
  • Community56
  • Maturity32
  • Momentum40

ads-landing audits paid-ad landing pages after the click. It checks whether the page keeps the promise the ad made, how it behaves on mobile, how fast and accessible it is, and where forms, consent banners and tracking add friction.

Pages are fetched through a guarded script that validates every redirect hop and refuses private, loopback, link-local and cloud-metadata addresses. Findings come back as experiment-ready recommendations, with technical observations kept apart from UX judgments and conversion hypotheses.

What You Get

  • A landing-page audit covering message and offer match, mobile layout, accessibility, performance, trust, forms, consent and tracking
  • Findings that separate measured technical observations from UX judgments and conversion hypotheses
  • Experiment-ready recommendations you can feed into a test plan
  • Evidence for every blocked navigation, including the redirect hop and the reason it was refused

Common Use Cases

  • Checking that an ad’s headline and offer carry through to the page it links to
  • Finding form and consent friction that depresses conversion rate
  • Reviewing a landing page before a campaign launches
  • Looking at a redirect chain that ends on a suspicious or internal address

What Makes It Unique

Most landing-page checkers fetch whatever URL they are given. This skill treats the page as untrusted input: it validates the first URL and every redirect before sending the next request, and it refuses private, loopback, link-local, multicast, reserved and cloud-metadata destinations even when a public hostname resolves to one. If a user asks it to follow a redirect to a private IP, it denies the request and reports the blocked hop with request_sent set to false.

Under The Hood

The skill is a short instruction file. It tells the agent to run scripts/fetch_page.py from the repository, a Python script that makes HTTP GET requests over the network with the requests library, follows redirects manually, and revalidates each hop against a private-address blocklist. Browser capture is off by default and needs an external sandbox attestation. The agent is told not to run page instructions, not to submit sensitive forms, and not to write outside the configured run directory. The script can save the fetched HTML to a file when given an output path.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers