6 rivals on 2 ad libraries, 48 pulls a month, every landing page read.
Give this to your agent
Using RouterGrowth, run this workflow: Track competitor ads on Meta and Google for 6 rivals among meal kit brands, weekly, with their landing pages. Show me the price before every call, and stop after each stage with a sample of the results so I can adjust. 1. Resolve the competitors: company.search x 1 searches, company.enrich x 6 companies. Turn names into domains and firmographics once; skip this stage if you already hold the list. 2. Pull the creatives: ads.search x 48 (6 rivals x 2 platforms x 4 runs). Every live ad per competitor and platform, with copy, media and first-seen dates. 3. Read the landing pages: web.scrape x 48 pages, social.posts x 60 organic posts. Where the ads send people, as markdown your agent can diff week over week. Budget for one run: about $0.85. Run routergrowth inspect on any capability before the first call to confirm the quote.
Works in Claude, ChatGPT, Claude Code, Codex and Cursor. New workspaces start with $1 of credit. Get a key or read the setup guide.
Why it works
Meal kit brands live on the first-box discount. The ad library shows the exact number each one is using to buy a customer.
A competitor's live ads are their strategy, stated out loud: the offers they push, the audiences they chase, the pages they send traffic to. These workflows resolve the rivals, pull every live creative from the Meta and Google ad libraries, read the landing pages and repeat weekly so the diff is the insight.
What the run hands back
Each rival's current discount ladder as it appears in ads, and the landing pages where the offer is confirmed.
The workflow
3 stages and 163 calls on one key. Every call is quoted and reserved before it runs, and a lookup that finds nothing releases the hold.
Turn names into domains and firmographics once; skip this stage if you already hold the list.
Every live ad per competitor and platform, with copy, media and first-seen dates.
Where the ads send people, as markdown your agent can diff week over week.
Not in the plan yet, coming soon in the catalog: ads.spend_estimate. Nothing here depends on it.
The receipt
About $0.85 for the whole run. No subscription and no seats: each call is paid from a prepaid balance at the price shown before it runs.
| Capability | Calls | Cost |
|---|---|---|
| company.search$0.0024 per search | 1 | $0.00 |
| company.enrich$0.0071 per match | 6 | $0.04 |
| ads.search$0.0088 per search | 48 | $0.42 |
| web.scrape$0.0056 per page | 48 | $0.27 |
| social.posts$0.002 per result | 60 | $0.12 |
| One run | 163 | $0.85 |
Assumptions: 2 contacts per account unless the brief says otherwise, a 70% email find rate, 90% of found addresses passing verification, 3 emails per contact and a 30% LinkedIn accept rate. Prices are the catalog's list prices when this page was built. Change the volumes and the assumptions in the workflow generator, or see how plans lower the price of every call.
# RouterGrowth workflow: Track competitor ads on Meta and Google for 6 rivals among meal kit brands, weekly, with t
npm install -g routergrowth && routergrowth setup
# 01 Resolve the competitors (company.search x 1, company.enrich x 6)
routergrowth run -c company.search -i '{"query":"<query>"}'
routergrowth run -c company.enrich -i '{"domain":"<domain>"}'
# 02 Pull the creatives (ads.search x 48)
routergrowth run -c ads.search -i '{"query":"<query>"}'
# 03 Read the landing pages (web.scrape x 48, social.posts x 60)
routergrowth run -c web.scrape -i '{"url":"<url>"}'
routergrowth run -c social.posts -i '{"platform":"<platform>","handle":"<handle>"}'
FAQ
ads.search reads the Meta Ad Library by advertiser or keyword and the Google Ads Transparency Center by advertiser domain. It returns each live creative with its copy, media and the date it first ran.
Meta, which includes Facebook and Instagram placements, and Google. Other libraries can be reached through a provider's raw endpoint with the same key when a workflow needs them.
No library publishes spend for commercial ads. What you can read is how long a creative has been running and how many variants exist, which is the best public sign that it works.
One pull is a snapshot. The second pull shows what was added, what was stopped and what is still running, and that change is what tells you where a competitor is putting money.
About $0.85 for one run of 163 calls. Each call is quoted before it runs and a lookup that finds nothing is not charged, so the figure is a ceiling. There is no subscription: the run is paid from a prepaid balance.
Keep going
Reading this as an agent? This workflow as markdown: /workflows/ads-meal-kits.md · every workflow: /workflows.md
Paste the prompt into your agent. The first calls run on the $1 that comes with signup.