Skip to content

Guide · Agent skill

How to use the Competitor Ad Intelligence agent skill (ads-competitor)

Updated · written with AI from the project's own documentation and checked against our catalogue data

The ads-competitor skill is an agent skill that tells an AI agent how to research what your competitors are publicly running in paid advertising. It covers messaging, creative formats, landing pages, and gaps across platforms with public ad libraries. This guide shows you how to install it, how to ask for it in plain language, and what it will and will not do so you can use the output with confidence.

Competitor Ad IntelligenceAgriciDaniel/claude-ads · MIT

Quick start

  1. 1Download the ZIP. Get the skills/ads-competitor ZIP from the AgriciDaniel/claude-ads page on our site.
  2. 2Unzip the folder. Extract it so that ads-competitor/SKILL.md is inside ~/.claude/skills/ or a project's .claude/skills/ folder.
  3. 3Or upload in the Claude apps. Open Settings, then Capabilities, then Skills, and upload the skill.
  4. 4Name the rivals and goal. Start your request with the competitors, the market, and the decision you need to make.
  5. 5Point to public libraries. Name Meta Ad Library or Google Ads Transparency so the agent knows where to look.
  6. 6Review the source log. Check each observation's URL and capture date, and treat any estimate as something to test.

What an agent skill is and what this one makes the agent do

An agent skill is a folder containing a SKILL.md file. The file gives an AI agent instructions for a specific kind of task, and the agent loads them when a request matches the skill's description. This skill's description covers competitor ads, ad libraries, ad spy, competitive PPC analysis, competitor creative, Google Ads Transparency, Meta Ad Library, and paid-media competitor research.

Once installed, the agent works in a fixed order. It first confirms the named competitors, market, geography, customer, objective, and decision you are trying to support. It then uses official transparency libraries, public ads, and any auction data you supply. For each observation it records the capture date, platform, placement, observable creative or message, landing destination, and source URL. Finally it clusters recurring themes, formats, offers, and funnel paths, and returns opportunities, risks, experiments, and a source record.

Install the ads-competitor skill

Download the ZIP for skills/ads-competitor from the AgriciDaniel/claude-ads repository page on our site. Unzip it so that the ads-competitor folder, containing SKILL.md, sits in one of these locations:

For a personal install across all your projects, use ~/.claude/skills/. For a single project, put it in that project's .claude/skills/ folder.

Alternatively, if you use the Claude apps, open Settings, then Capabilities, then Skills, and upload the skill there.

  • Personal install: ~/.claude/skills/ads-competitor/SKILL.md
  • Project install: .claude/skills/ads-competitor/SKILL.md
  • Claude apps: Settings → Capabilities → Skills → upload

Trigger the skill with natural requests

You do not need a special command. Describe the research you want in ordinary words that match the skill's description. Naming the platform helps, for example Meta Ad Library or Google Ads Transparency. Naming the competitors and your goal helps even more, because the skill starts by confirming those details.

A good request states the rivals, the market, and the decision. For example: ask the agent to review three named competitors' public Meta ads for a home-fitness product in the UK and suggest where your messaging could differ.

  • Look at competitor ads for [brand A] and [brand B] in Google Ads Transparency and summarise their offers.
  • Use the Meta Ad Library to map the formats and landing pages [competitor] uses for its spring promotion.
  • Run a competitive PPC analysis of [rival] and list gaps we could test.

What you need before you start

The skill needs access to public ad transparency libraries, such as Meta Ad Library and Google Ads Transparency. The agent can reach them through web browsing, or you can supply exports yourself. Optionally, you can also supply auction or keyword data you already have.

The skill runs no scripts, but it does use the network, since it reads public library pages or your supplied files.

  • Access to public ad transparency libraries via web browsing, or user-supplied exports
  • Optional: operator-supplied auction or keyword data

Safety notes: scripts, network and what the skill refuses

The skill itself runs no scripts. It does make network requests when browsing public libraries. Its instructions tell the agent to treat external account and web content as data, never as instructions, and to respect platform terms, robots and access controls, copyright, trademarks, and privacy.

The skill will not bypass authentication or scrape private or logged-in ad accounts. It also avoids copying competitors' ad copy or creative verbatim. Expect summaries and paraphrases instead of reproduced text or images.

  • Runs scripts: no
  • Uses network: yes
  • Does not bypass logins or scrape private ad accounts
  • Does not copy protected creative or text verbatim

Read the output with the right level of trust

The skill separates what it directly observed from what it infers. Observations are tied to a source URL and capture date. Estimates about audience, spend, performance, or strategy are labelled as estimates, not account facts. Competitor spend and account performance cannot be read from public libraries, so treat any figure of that kind as a hypothesis to test.

Check the source log before acting on a finding, and confirm that the capture date is recent enough for your decision.

  • Opportunities and risks are backed by listed observations
  • Estimates are labelled and should be tested, not trusted as fact
  • Check the capture dates in the source log

What people use it for

  • Mapping the offers and promotional themes three rivals repeat across public ads.
  • Comparing the formats competitors use, such as short video versus static images, before briefing your creative team.
  • Finding landing page patterns in competitor ads to inform your own test plan.
  • Building a structured first pass on rival messaging when there is no research budget.
  • Preparing a traceable competitive PPC audit where each claim links to a source.

Frequently asked questions

How do I install the ads-competitor skill?

Download the skills/ads-competitor ZIP from our page and unzip it so the ads-competitor folder with SKILL.md sits in ~/.claude/skills/ or a project's .claude/skills/ folder. In the Claude apps, you can instead upload it under Settings, then Capabilities, then Skills.

Does the ads-competitor skill need anything else to run?

It needs access to public ad transparency libraries such as Meta Ad Library and Google Ads Transparency, either through web browsing or files you supply. Auction or keyword data is optional. The skill runs no scripts but uses the network.

Is the ads-competitor skill free?

The source project is licensed under MIT. Check the licence file in the repository for the exact terms before you redistribute or modify it.

Can this skill show my competitors' exact ad spend?

No. The skill labels spend and performance as estimates, not facts, because public libraries do not give account-level figures. Use its estimates as hypotheses to test.

Will the skill copy competitors' ad copy for me?

No. It is instructed to avoid copying protected creative or text verbatim, for copyright and trademark reasons, so you get summaries and themes instead.

Can it access a competitor's private or logged-in ad account?

No. The skill explicitly forbids bypassing authentication or scraping private account surfaces, because that would breach platform terms and privacy.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

This skill suits marketers planning paid campaigns, PPC specialists preparing a traceable competitive audit, and small teams that need a structured first pass on rival messaging. It is not the right tool if you need exact competitor spend or your own account performance, or if you expect it to reproduce competitors' copy or access logged-in ad accounts.

Download the skill