Marketing systems

Competitive ads intelligence

An agent that reads the LinkedIn, Meta, and Google ad libraries and writes a weekly report on what competitors changed.

What it does

The agent starts from your company, finds your closest competitors, pulls their live ads from the three public ad libraries, compares them against last week, and writes a branded PDF. A schedule runs it every Monday morning.

High-level diagram: one command in, three ad libraries read, branded PDF out

Three ad libraries, one branded report, on a schedule

How it works
Finds the competitors. Checks G2, TrustRadius, and category roundups, with evidence for each pick.
Captures the three ad libraries. Screenshots plus parsed ad copy and exact counts for every company.
Compares against last run. New ads, dropped ads, volume shifts — launches separated from ad rotation.
Writes the report. A two-page brief, one page per competitor with ad quotes, evidence in the appendix.
Saves what it learns. Workarounds discovered during a run are saved and read at the start of the next one.
Step-by-step trail of one unattended Monday run
Example from a run
Your competitor went from 1 to ~150 Meta ads in five days

From a run on Asana’s market: the report flagged monday.com’s jump — 78 new creatives — while its LinkedIn and Google ads stayed flat. A campaign launch, caught the week it started.

12/12ad-library captures, unattended
~17 minfrom schedule firing to finished PDF
2 pagesof executive brief, evidence behind it
Report example
First page of the report

The full PDF from the Asana run — the two-page brief, one page per competitor, and the evidence appendix.

Download the report
See the other builds

Automated style guide

Writes your style guide from your published articles and reviews every new draft against it.

View the build →

Pricing teardown

Grades pricing pages against a pricing expert’s rubric and ranks a whole market in one table.

View the build →

Adapted from the competitive-intelligence agent published by Kamil Rextin of 42 Agency. This rebuild adds local infrastructure, parsed ad data with exact counts, change tracking between runs, and a lessons loop that makes each run smarter.