agentleFS
EricStrohmaierEricStrohmaier/adchat-mcp/skills/cut-wasted-spend/SKILL.md

cut-wasted-spend

Skill1 starsChanged 3 days agoMITView on GitHub

Find and stop Meta and Google Ads spend that isn't producing results. Use when the user asks to cut waste, reduce costs, find what's not working, "where am I losing money", pause bad ads, lower budgets, or clean up an account in Meta (Facebook/Instagram) or Google Ads through AdChat.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Not waste: anything under ~3 days old, in learning, or with too little spend to judge.

    line 24
  2. Distinctive:

    Don't pause the last active ad in an ad set without saying the ad set will stop.

    line 72

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Cut wasted spend

Goal: stop paying for things that have had a fair chance and failed, and keep everything that's still learning or working. Pausing too early wastes money too.

What counts as waste

Over the last 14 to 30 days:

  • Spent 2–3× the target cost per result with zero results. Target from get_account_briefing; if there is none, ask.
  • Cost per result 2× target or worse with enough spend to judge.
  • Audiences or placements (Meta breakdowns) eating budget with no results.
  • Search terms that cost money and will never convert (Google; see google-search-cleanup).
  • Duplicate effort: two ad sets or campaigns targeting the same people and bidding against each other.

Not waste: anything under ~3 days old, in learning, or with too little spend to judge.

Steps

  1. Context. get_account_briefing for targets and notes ("don't touch brand campaign").
  2. Meta.
    • meta_insights at level: ad for the last 14–30 days, sorted by spend. Flag ads with spend ≥ 2× target and no results, or cost per result ≥ 2× target.
    • Same at level: adset.
    • meta_insights with breakdowns: ["publisher_platform", "platform_position"] and ["age"] on the biggest spender for segments with spend and no results.
  3. Google.
    • google_list with level: campaigns, then level: keywords (cost, conversions, quality score).
    • google_list with level: search_terms: costly searches with no conversions.
    • google_get_insights with breakdown: network and device for leaks such as display or search partners.
  4. Check alerts. get_alerts may already list waste with suggested changes.
  5. Propose. One table per platform, biggest saving first.
  6. Apply after the user picks. Pausing and lowering budgets need no Yes on autopilot, but always show the list first and let the user remove items; on review mode ask Yes/No. See safe-changes.
    • Meta pause: meta_update with status: PAUSED, or meta_bulk_update with preview: true first for several.
    • Meta budget down: meta_update with the new daily_budget in cents.
    • Google pause: google_set_status with status: PAUSED.
    • Google negatives: google_add_negative_keywords (google_add_keywords adds paid keywords).
    • Google budget down: google_update_campaign_budget in currency units.

What to show the user

Waste found · last 30 days · target CPA $25 · about $640/month

Meta
Ad / ad set                     Spend   Results  Why                      Action
Static B (Broad US)             $210    0        8× target, no results    Pause ad
Audience Network (all ad sets)  $95     0        no results               Exclude placement*
Google
"free jacket pattern"           $48     0        no buying intent         Negative (phrase)
Display expansion, Campaign A   $130    1        $130 per conversion      Discuss

*Placement edits change targeting on an ad set: name the ad set and the change.

Guardrails

  • Pausing is reversible; deleting is not. Prefer pause.
  • Don't pause the last active ad in an ad set without saying the ad set will stop.
  • Editing an ACTIVE ad set's targeting needs a Yes on autopilot.
  • Meta budgets in cents; Google in currency units.
  • Be honest about uncertainty. "Probably waste" items go in a "discuss" column.

Example prompt

"Find where I'm wasting money on Facebook and Google this month and pause the worst."

Briefing → Meta ad and ad set insights → breakdowns → Google campaigns, keywords, search terms → table → user picks → pauses and negatives → one-line summary per change.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.